Reindexing old data automatically

Hello all,

I need to keep my granularity to “hour” for data older than a week and “minute” otherwise.

Can I automate this?

I know that using “dataSource” input spec (since I use hadoop based ingestion) I can reindex old data, but will it automatically reindex as data gets older? Or will I have to submit a new “dataSource” index task repeatedly?

Hey Rishi,

There is no builtin way to automate this, so periodically submitting reindex tasks is the way to go.