Starting to move the analytics DDB from the data accounts into this stack. Reasoning being: this "temporary cache of DTR redirect data" should be tied to a specific region/Dovetail-stack.
So this PR:
Creates a new table directly in the dovetail-analytics.yml stack
Sets up capacity scaling (hopefully) the same way as the data accounts
Adds a temporary redundant analytics-dynamodb-lambda function, hooked up to the same kinesis stream as the real thing.
New lambda will insert (hopefully) the same data into the new table. But has no output subscription-filter.
In > 24 hours (hopefully) our tables will reach parity. And we can point the real function at the new table, and delete this temporary function.
Starting to move the analytics DDB from the data accounts into this stack. Reasoning being: this "temporary cache of DTR redirect data" should be tied to a specific region/Dovetail-stack.
So this PR:
dovetail-analytics.ymlstackanalytics-dynamodb-lambdafunction, hooked up to the same kinesis stream as the real thing.