[Bug] dbt source freshness doesn't care require-dbt-version with 0.21.0

[yu-iskw]

Is there an existing issue for this?

• I have searched the existing issues

Current Behavior

Even if we specify older version(s) with require-dbt-version, dbt source freshness ignores the version constrain.

Expected Behavior

dbt source freshness should not work with older version constrain.

Running with dbt=0.21.0
Encountered an error while reading the project:
  ERROR: Runtime Error
  This version of dbt is not supported with the 'xxx' package.
    Installed version of dbt: =0.21.0
    Required version of dbt for 'xxx': ['=0.19.0']
  Check the requirements for the 'xxx' package, or run dbt again with --no-version-check

Steps To Reproduce

1. Install dbt 0.21.0
2. Specify older version with require-dbt-version: "0.19.0"
3. Execute dbt source freshness

Relevant log output

No response

Environment

- OS: maxOS Big Sur
- Python: 3.8.8
- dbt: 0.21.0

What database are you using dbt with?

bigquery

Additional Context

No response

[jtcohen6]

Thanks for the report @yu-iskw. In all dbt versions up to v0.21, there is a lot of inconsistency in which commands used/respected require-dbt-version + --no-version-check.

For v1, we reworked all "global configs" (#3839), including VERSION_CHECK, so that now:

• All commands will adhere to version checks
• Version checking is on by default
• You can turn version checking off via CLI flag (--no-version-check), env var (VERSION_CHECK=False), or profiles.yml config

Prerelease docs on this: https://next.docs.getdbt.com/reference/global-configs#checking-version-compatibility

I'm going to close this issue, since I believe it's resolved in the next version of dbt-core