Hi all / Hong.
Slight contradiction picked up, currently Flink CDC 3.4 (Postgres) is not
compatible with Flink 2.x
https://nightlies.apache.org/flink/flink-cdc-docs-release-3.4/docs/connectors/flink-sources/overview/#supported-flink-versions
So the setting above wont make a difference.
G
On Thu
Hi Wang
As per previous, this is already how my CDC is configured.
any other ideas.
here is my table create.
CREATE OR REPLACE TABLE postgres_catalog.inbound.adults (
id BIGINT -- This is a postgresql Serial generated field
,nationalid VARCHAR(14) -- NOT NULL
,data STRING -- JSONB Payload
,creat
Flink 1.20.x does not allow me to unpack/repack the json as a string into
the array of rows as a complex structure...
my inbound json payload is packed into a column called "data" as type
jsonb...
with the 1.20.x functionality I can't unpack jsonify it... and the
functionality needed is only in 2
Hi there
Had a look was/is using this setting on the CDC source, see below.
CREATE OR REPLACE TABLE postgres_catalog.inbound.adults (
id BIGINT -- This is a postgresql Serial generated field
,nationalid VARCHAR(14) -- NOT NULL
,data STRING -- JSONB Payload
,created_at TIMESTAMP_LTZ(3)
,WATERMARK
Thank you.Will give this a try and see if using 2.0 gives me the extra functionality to unpack my payload. Although if I follow the docs the extra functionality is only part of 2.2.Sent from my iPhoneGeorge Leonard__george...@gmail.com+27 82 655 2466On 18 Sep 2025, at 03:55,
Hi George,
Please use ,'scan.incremental.snapshot.enabled'= 'true' . The old
SouceFunction has been removed in flink 2.0. Otherwise, you can use flink
1.20.
Best,
Hongshun
On Sun, Sep 14, 2025 at 11:48 PM George wrote:
> Hi all...
>
> Below is the jars included in my flink 2.0 build. and th
