I’m exploring the possibility of syncing a PostgreSQL table to ClickHouse using the VersionedCollapsingMergeTree engine as the destination. While testing, I observed that:
The PeerDB GUI currently supports only MergeTree and ReplacingMergeTree engines.
I manually created a ClickHouse table using the VCM engine with _sign and _version columns, but the mirror validation failed with:
"CDC Mirror is invalid: not all PeerDB columns found in destination table".
Could you please clarify the following:
- Is it possible to use
VersionedCollapsingMergeTree as the destination engine in PeerDB, either via GUI or custom config?
- If so, how can we map the necessary CDC fields like _sign and _version during sync setup?
- Is there a recommended approach to integrate
VCM cleanly with PeerDB’s CDC mechanism?
Any guidance or documentation you could provide would be very helpful.
Thanks,
Mayuresh Ovhal.
I’m exploring the possibility of syncing a PostgreSQL table to ClickHouse using the
VersionedCollapsingMergeTreeengine as the destination. While testing, I observed that:The PeerDB GUI currently supports only
MergeTreeandReplacingMergeTreeengines.I manually created a ClickHouse table using the VCM engine with _sign and _version columns, but the mirror validation failed with:
"CDC Mirror is invalid: not all PeerDB columns found in destination table".
Could you please clarify the following:
VersionedCollapsingMergeTreeas the destination engine in PeerDB, either via GUI or custom config?VCMcleanly with PeerDB’s CDC mechanism?Any guidance or documentation you could provide would be very helpful.
Thanks,
Mayuresh Ovhal.