What is the problem?
The Site Reliability Team, with the assistance of many other affected teams in the organization, will be performing a data center switchover from our primary data center (eqiad) to our secondary data center (codfw) for MediaWiki and other associated services, and a switch back to the original state, 3-4 weeks later.
The problem is that this is user-visible, as all wikis will need to be made read-only on both occasions for a period of roughly 10 minutes.
How can we help you?
As in previous years, we will need to communicate the upcoming downtime for both occasions to our users/communities, and explain why we are doing this.
What does success look like?
Success would look like awareness by all regular users of our wikis that this will happen and that they won't be able to make any changes for a short period of time (twice), and, as in recent years, some understanding by our users why this maintenance is important in protecting our sites and services with disaster readiness. We were very happy with how Community Relations has handled this project in the past 2 years and are looking for a very similar result.
What is your deadline?
Tentative timeplan currently is. Note that only the mediawiki part is expected to have visible user repercussions as the wikis will need to be in a read-only mode for a while.
Switchover
Services: Tuesday, September 11th 2018 14:30 UTC Media storage/Swift: Tuesday, September 11th 2018 15:00 UTC Traffic: Tuesday, September 11th 2018 19:00 UTC Mediawiki: Wednesday, September 12th 2018: 14:00 UTC
Switchback:
Traffic: Wednesday, October 10th 2018 09:00 UTC (and maybe some prep work on Monday) Mediawiki: Wednesday, October 10th 2018: 14:00 UTC Services: Thursday, October 11th 2018 14:30 UTC Media storage/Swift: Thursday, October 11th 2018 15:00 UTC
We should probably be alerting at least 2-3 weeks beforehand