Andy Asp
Andy Asp
Oh whoops, assigned myself before reading your latest comment. I'll leave this be then.
This may be because there is a [minio](https://github.com/grafana/mimir/blob/4de8352265b7c26494ba32923bfc932ce07bf809/operations/helm/charts/mimir-distributed/values.yaml#L2321) section in the values file. If you're setting up your minio seperately then at the top-level you could counter-intuitively set ```yaml minio:...
> It seems like some policies are not set properly. I'm not sure what this means, but it sounds like maybe you figured it out? > Is there a guide...
Interesting. I don't think that's a bug with the sanity check. It's rather simple and creates the bucket client and GETs a single object ([code reference](https://github.com/grafana/mimir/blob/092aee1000b935b99f6a60024251409e86daae88/pkg/mimir/sanity_check.go#L147)). Here the error it's...
> they treat all responses with 404 as "bucket not exists" That isn't true and would be very incorrect behavior. Requesting an object that doesn't exist results in 404 and...
As an update to the current state of this: I've gone through to address most non-test comments. My focus will now be on fixing up the remaining test comments. There's...
As a note, the test changes are still a work in progress. The next step will be to coordinate in the test when the async validation has stopped (hopefully in...
Coordinating the async validation in the tests is now done. Barring new discoveries, remaining work to get this ready for review again is to double check what I promised to...
@pstibrany Thanks for the review. > I would strongly suggest to not add testing of validation into TestMultitenantCompactor_FinishBlockUpload. That test covers functionality of FinishBlockUpload. Validation is a new concern and...