Debating Office 365 Robustness and Stability


The Cloud, The Cloud, and Nothing but the Cloud

As promised, yesterday Mary-Jo Foley and I met up to discuss the topic of “How Stable is Office 365 – can I trust the cloud.” You can read the transcript of the discussion or listen to the recording online.

I hate reading transcriptions. Unlike writing, you don’t get the chance to smoothen and refine what you say unless you are very careful in how you speak. And when you speak with someone you know and trust, your speech tends to become a little less formal and unstructured. At least, that’s my only excuse for what appears in the transcript. (Note to self: Do I really say “Uh” and “Um” that much?).

Topics for Debate

In any case, we covered topics like:

  • The nature of incidents within Office 365 and the fact that something is going wrong somewhere at any time.
  • How incidents impact the Office 365 SLA and the scale of an incident necessary to move the SLA needle.
  • How to find out what’s going on when an Office 365 incident is happening.
  • Why it sometimes takes time before Microsoft acknowledges that an incident is real and sources to consult to find out what’s going on (the missing reference is to
  • Why it’s impossible to restore service with backups (because there’s nowhere to restore the data to).
  • The difficulty of moving data from Office 365 to another cloud service (like G Suite) or back on-premises.
  • The investment Microsoft has made in Office 365 datacenters and how this helps resilience by limiting problems to within one region (mostly).
  • The kind of catastrophe it would take to knock an Office 365 region out.
  • How my Twitter handle came about…

The time went incredibly quickly and I enjoyed the conversation. I hope it’s the first of many MJFChats I get the chance to be part of.

I acknowledge that the opinions are mine and only represent my views on the topic. If you want to contribute to the debate, do so in the forum. You never know, Mary-Jo might just make you her next guest/victim.