API Error: 400 - group memberDivisions count exceeds the limit of 50

When trying to deploy PR to have member_division_ids = ["*"] as list of Divisions to associate to the SkillGroups, we are getting the below error:

Error: Failed to create/update skill group RES1_xxx_AG member divisions: API Error: 400 - group memberDivisions count exceeds the limit of 50

Has anyone experienced this error in their environment?
How did you resolve the issue?

Thanks

Hi @Nishant_Tank

It is a product limitation essentially.
You can request changes to this behavior and share your use case at https://purecloud.ideas.aha.io/ideas

Thanks
Hemanth

Hemanth
We have tested using "*" for SkillGroup division property within API Explorer and IT WORKS.
This feature was requested by Airbnb, and it has been implemented in Genesys Cloud.
But it is not working via terraform. So as a last step, we need Genesys to enable this feature (which is already working otherwise), to also work when being requested through terraform.

Regards
Nishant Tank

Hi Nishant,

Please work with your TAM and we can work with him to get a ticket in our backlog. I can't comment on this feature's specifics, but we have had situations where the product team implements a new feature and the change is not communicated to the Cx as Code team. So if you work with your tam (Premkumar) we can track it and let you know when we can start work on it

Thanks,
John Carnell
Director, Developer Engagement

Hi @Nishant_Tank ,
I am one of the developers of the Cx as Code team working on this issue. Can you provide the Correlation ID along with the full error message so that we can track this down?

Thanks,
Shruti

Hi Nishant,

We have not heard back from you in a few days. Can you please re-run your code and send via your tam the full error message along with the correlation id (it will be the GUID in the message). Also, sending us the resource you are executing would be helpful. We have been trying to reproduce the error in our lab and talking with the developers who own the service. We need the guid to nail down the specific call and see what is going on here

Thanks,
John Carnell
Director, Developer Engagement

Hi @John_Carnell
Apologies for delay in response. Been tied up with some other ongoing stuff.

I have passed the requested details to Premkumar.
However, I could not re-run the code so have sent the details from the run that had failed.
Let me know if those details are still useful or if I definitely need to re-run the code.

Regards
Nishant Tank

Hi Nishant,

I will look at the data you sent us and lent you know what is going on. We have a dev working on it and the service team is involved so hopefully we can get to the bottom of it.

  • John

Hi Nishant,

I think I found the problem. It has to do with updating the skill group regarding member division. When a skill group is updated, it pulls back a list of all the member divisions the skill group has assigned. If the member divisions were originally set with individual lists of division IDs (for example by hand in the UI or via a script or terraform) and then you try to just update using a * the API is going to try and take the items previously assigned and then add the * parameter to the list. The API still enforces a maximum limit of 50, so when you try to update any kind, the call will fail.

I have been talking with the API team about how we can improve this, but in the meantime, I am going to have @shrutisuryawanshi modify the update method in the CX as Code resource so that if the CX as Code resource definition contains a * in the list of resources, it will automatically remove all the previous definitions and replace them with a *. The API enforces either 50 resources or you can assign all, but it does not attempt to be exclusive or validate one or the other.

I cant give an ETA yet on this but I am hoping we can get this resolved in the next of CX as Code resource.

Thanks,
John Carnell
Director, Developer Engagement

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.

Hi Nishant,

This was fixed in release v1.54.0 (our last release). I apologize; I forgot to ask the developer to update the forum post with a note saying it was released.

Thanks,
John Carnell
Director, Developer Engagement

2 Likes