We have it figured out, it was DNS, and we've done many migrations of mail servers with the other cluster, and even this one has had some.
Of course it would not be a bad idea to check in on things over the weekend, but we will be aware and looking for this and if it occurs resolve it much more quickly than before.
Last time we had a temp fix, then worked with Parallels toward a global fix, which brought its own problems that we came up with an internal solution on as Parallels simply takes too long in such cases to make real solutions. We wanted to do the changes with their 'blessing' and support in future, and attempted that route, but when things don't go as planned you have to do things other ways. Again, we'll be keeping an eagle eye on it and keep updated if there is any issue via here and twitter.