@Gareth Oates Sharing as answer based on response I got from Product Group:
There was a repair on 2/2 which created a new pod with Ip 10.24.2.5.From the job traces it seems repair completed with success, the old cg may have took a while to clean up but I don't see much of a reason how the new one failed to be connected with since the consistency job completed with no issues, I think the only way to find out is for user to keep their CGs in running state so we can take a closer look.
I can see on another thread opened by you that you are working with Support so they can help you better out with issues since they do have required tools and expertise to help you out on issue.
Hope it helps!!!
Please "Accept as Answer" if it helped so it can help others in community looking for help on similar topics.