== This is (almost) the final step! ==
Progressively move the remaining 30% of external traffic.
**This means all external API traffic, all web browser, all mobile application traffic, all internal traffic, and all jobs will be served by #mw-on-k8s**
=== Exceptions ===
==== External traffic ====
- [] Commons, pending {T295007} and {T295007}
- [] Votewiki, pending {T209892}
- [] Wikitech, pending {T292707}
- [] Dumps
==== Jobs ====
- [] Videoscaling jobs, pending {T355292}
==== Internal traffic ====
- [] Mediawiki calling itself, pending Commons work
- [] ml-serve services {T362316}
- [] Possibly yet-unidentified stragglers
== Progression ==
[] 75%
[] 80%
[] 85%
[] 90%
[] 95%
[] 100%
Info from {T351074}:
> For every 5% of external traffic we move, we've needed to bump mw-web by 12-13 replicas and mw-api-ext by 10 replicas.
>
> This means that for every 5% increase in traffic, we're requiring 22-23 additional replicas. Given every pod requires 5.6 CPUs it means we're going to need about 123 cores per traffic bump, or roughly 3 servers as our servers have 48 cores each.
>
> The above calculation is per-datacenter, of course.