CloudQuery config changes causing duplicate key error with deterministic_cq_id and write_mode

I recently added deterministic_cq_id: true to my config and write_mode: overwrite and am now seeing errors:

"err": "ERROR: duplicate key value violates unique constraint \"aws_lambda_runtimes__cq_id_key\" (SQLSTATE 23505)"

Is there a way to make the change transition smoothly, or what am I misunderstanding about this error?

So typically when users use deterministic_cq_id: true, they also use pk_mode: cq-id-only. This enables the database to only have to care about a single PK during insert.

CloudQuery Documentation: pk_mode

What database and plugin are you using?

Does adding that flag affect the PK of the table, or does it simply change the CQID generation method?

Postgres DB
And AWS source

It impacts the PKS on the table… Rather than having a composite PK, it uses a single column _cq_id.

Are you getting that error on every single sync?

Yes, I am.

Well, so I’m talking specifically about the deterministic cqid and omitting the cq-id-only flag.

Ok. Can you open an issue on GitHub with the details of your config and we will look into it!

Open an Issue on GitHub

Yes, the deterministic_cqid flag will only change how the _cq_id is calculated; otherwise, it will be a random value.

So it shouldn’t impact the primary key like it appears to be.

Actually, I think that I understand the issue. Some deterministic UUID must be colliding with a previously random UUID?

Is there an overwrite flag for such a situation?

If I have overwrite delete stale on the Postgres spec, will those entries get deleted and then recreated?