Office 365 gotcha No. 3: Failing to ensure prerequisites are met for hybrid deployments
You've done your due diligence on your design decision and have decided the hybrid approach is best. You're going to blend some of your existing on-premises Exchange instances with Office 365 accounts. But there's a catch. You may not be able to connect to Office 365 from your on-premises environment if you don't have the prerequisites in place.
Hybrid deployments can be configured with Exchange 2007 through 2013 environments. However, with Exchange 2007/2010 environments there must be at least one Exchange 2013 Client Access and Mailbox server in place to run the Hybrid Configuration wizard. It's recommended you run the two roles on the same server. If your organization doesn't have Exchange 2013 (CU1 or higher) in place, you'll have to update your environment before going hybrid.
Office 365 gotcha No. 4: Choosing the wrong support plan
Office 365 support plans are by no means equal. Larger organizations are unlikely to get tripped up by this gotcha, but smaller shops that require a hybrid approach beware: It's essential to have a plan in place that supports Azure Active Directory synchronization if you're going hybrid. All of the Enterprise, Government, Academic and Midsize plans support Azure AD sync; as a result, they also support hybrid deployment. The Home and Small Business flavors of Office 365, however, do not offer Azure AD sync, and therefore do not support hybrid deployments. Don't be penny-wise, pound-foolish.
Office 365 gotcha No. 5: Getting derailed during legacy archive export
Most on-premises legacy archive solutions won't work with Office 365. You move the mailbox to the cloud, and it breaks the stubs, leaving users unable to access archived email. There are a variety of ways to get your data out of your legacy archive solution. Your archive solution might have export capabilities to help extract the data, but the benchmark speed for the export might be 1GB to 2GB per hour (yikes). In this case, you would be exporting the data to PSTs to then upload to Office 365. This may be a real problem — you'll be rehydrating the data up to its original size because you won't be able to use deduplication. You will also have problems if your archive's databases are corrupt, because the legacy archive API may not handle unhealthy indexes or databases well.
A second approach is to use the native export tools in the legacy archive solution to rehydrate the stubbed messages back into the mailbox directly (this works for mailbox archives but not journals). Again, the rehydration process can be a real issue because you'll need to allocate however many terabytes of storage (depending on your data size) or rehydrate one mailbox at a time.
Sign up for CIO Asia eNewsletters.