DanielLeens commented on PR #12015:
URL: https://github.com/apache/seatunnel/pull/12015#issuecomment-5696080027
@luozihen Good diagnosis on your part — you've already found the right
error, and it's not something in your control to fix by changing test logic.
Confirmed it directly from the fork's run logs, so here's the full picture
rather than a "just retry" answer.
**Root cause, confirmed from logs (not a guess):** Docker Hub's
`minio/minio` repository was removed entirely back in mid-September — it's a
404 `pull access denied ... repository does not exist`, not a rate limit or
transient network blip, and it will never succeed no matter how many times the
job is rerun. I pulled the actual failed-job logs from your fork's run
(`luozihen/seatunnel` run `34802808218`, run-attempt 12) and confirmed the same
signature on all 8 failing legs:
```
Status 404: {"message":"pull access denied for minio/minio, repository does
not exist or may require 'docker login': denied: requested access to the
resource is denied"}
```
— hit by `PaimonWithS3IT.startUp` (`paimon-connector-it`), `DatabendIT` /
`DatabendCDCSinkIT` (`all-connectors-it-2`/`-6`/`-7`), and the same pattern on
the other failing legs — every one of them a test that spins up a MinIO
Testcontainer, unrelated to the actual test assertions.
**This is not caused by your diff.** Your PR only touches `seatunnel-api`,
`connector-jdbc`, `seatunnel-core-starter`, and docs — none of the failing test
classes (Paimon/Databend/S3 E2E) are anywhere in your changed-files list, so
there's nothing in your JDBC primary-key-mapping change that could produce this.
**It's already fixed on `dev`, your branch just doesn't have it yet.** This
exact issue hit a large batch of PRs a few days ago and was fixed in
apache/seatunnel#12287 (merged 2026-09-13), which swaps the image reference to
`quay.io/minio/minio` (Docker Hub's `minio/minio` repo is gone for good, not
coming back). Your branch head (`2c5982be8`) is currently 174 commits behind
`dev` — it predates that fix, which is exactly why every rerun keeps failing
the same way.
**What to do:** no code change needed on your side, and no need to touch
your actual feature logic — just pick up the fix from `dev`:
```bash
git fetch origin dev
git rebase origin/dev
# resolve any real conflicts (unlikely — the fix only touches unrelated E2E
test files)
git push --force-with-lease
```
That should bring in the `quay.io/minio/minio` swap and clear the
Paimon/Databend/S3 failures. Ping me once CI's green again on the new head and
I'll take a fresh look.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
