We adopted HVR in 2019 after becoming very disenchanted with the Attunity (now Qlick) product.

We have been very happy with the HVR product and have invested significantly in it over the past 5 years.  We now have multiple source endpoints (Oracle, MS SQL Server, Postgres) and targets (Kafka, S3, Postgres) and our team has deep product knowledge and have completed the migration of mutiple existing HVR 5 channel configurations to HVR 6 over the past year and a half.

We now have a need to support MongoDB as a source (we are likely to have multiple MongoDB endpoints added over time).  However, we were surprised and disappointed to learn that HVR only supports MongoDB as a "target" and not as as a "source" endpoint. 

We have just spent the last few weeks (still in-progress) conducting a Fivetran-Hybrid POC for a MongoDB source.  Currently the Fivetran-Hybrid product doesn't support S3 as target. Furthermore, it is doubtful the new S3 connector once developed will be able to generate and deliver .csv (gzipped) files to S3 in exactly the same way that the HVR product does today. 

We have our own inhouse CDC framework that is designed to work with HVR generated S3  (gzipped) .csv files.  We would like to continue to use this inhouse CDC framework to ingest MongoDB CDC events.  But again, it doesn't look like we will be able to do this using the Fivetran-Hybrid product.

We have come to the conclusion that it would be better for us if the HVR product added MongoDB as a "source" endpoint.  We would HIGHLY prefer this approach to using the Fivetran-Hybrid product which is completely different; requiring different infrastructure, with different CDC event outputs, and a completely different set of support knowledge and skills.

As of late, it feels like the HVR product roadmap is being neglected.  We'd like to see Fivetran/HVR commit to ongoing investment and development of the HVR (LDP) product.  Please give this request serious consideration.

Steve Brackenbury
Senior Systems Architect
Teranet Inc.