missing information about s3 usage

Docs don’t explicitly say that when one is trying to use s3 or s3 compatible storage there should already exist bucket.

Then, if you would try to push data to s3 and bucket wouldn’t exist, you would get the following:

Preparing to upload data to 's3://my/remote/'
Preparing to collect status from s3://my/remote/
[############                  ] 40% Collecting information
ERROR: failed to push data to the cloud - An error occurred (NoSuchBucket) when calling the ListObjectsV2 operation: Unknown

So I propose to explicitly say in docs that you should already have existing bucket to be able to push to s3. Following docs are related to s3 usage: https://dvc.org/doc/get-started/configure https://dvc.org/doc/commands-reference/remote-add Related discussion in ods: https://opendatascience.slack.com/archives/CGGLZJ119/p1559831715009900

[shcheklein]

@Naba7 definitely, let’s mention this!

[dnabanita7]

Actually, it is mentioned in use-cases to create a bucket before connecting it to dvc remote as of here https://dvc.org/doc/use-cases/share-data-and-model-files . But I think mentioning it in commands-reference/remote and commands-reference/config will do good.