XDCRs and TTLs


#1

Hi

1.I would like to know if theres a wahy to define different TTLs between XDCR replications

For instance - we have an online bucket with a certain TTL
We replicate the documents the documents to a different cluster for warehousing and analytics - and I would like the data on the warehouse to be without a TTL

Is this a possible configuration?

  1. Is there a way to get notifications (via Webhooks for instance) when a document has reached its TTL limit?

Thanks
Arik


#2

Hey Arik,

At the moment, there isn’t a way to abstract the TTL to only 1 cluster.
One workaround would be to duplicate the data in 2 buckets, 1 bucket with data that has TTL’s , another that doesn’t. Replicate the bucket that doesn’t have TTL’s and then periodically drop and recreate it.

Another way some users manage expiry is use multiple buckets in a round robin format ever day, at the end of the day they switch to a new bucket, the older bucket is then deleted and recreated. You can delete the bucket when the other side is caught up. Then reestablish replication when you recreate the bucket.

hope this helps.