Planned downtime while services move to new hardware

Over the coming weeks the platform moves off VMware and onto a new virtualisation platform. The three physical servers are converted one at a time, and each one has to be emptied, rebuilt and refilled before the next is touched.

What this means for you

• Services will be unavailable during scheduled windows, not continuously.
• Every window is announced before it starts.
• The largest window is the final server, which carries most of what you use day to day. Expect that one to span more than a single evening.
• Nothing is being removed and no data is being discarded — this is the same services on different hardware.

Why now. One of the three servers has storage that has repeatedly failed under load and has already caused outages. Fixing it means rebuilding that machine, and rebuilding it on the platform we are leaving would be work thrown away.

Follow this post for window announcements.