Skip to content

Endless spinner when opening a large s3 bucket with tons of versioning #18020

Description

@ayounes2

Describe the bug
When trying to connect to an s3 bucket that has versioning enabled, cyberduck attempts to grab all the files and their historical versions using the S3 List Object Versions API instead of the regular List Objects API.

That API returns every historical version of every file, so instead of getting back "here are 5 items at the root", S3 was returning thousands of records, paginated 1,000 at a time. Cyberduck had to exhaust all its pages before it could display anything.

To Reproduce
Steps to reproduce the behavior:

  1. Go to an s3 bucket with versioning enabled
  2. Have the files in there change tons of times (100k+ versions)
  3. Try opening the bucket

Expected behavior
The default should be to call the regular object list api instead of versioning.
That config already exists in /cyberduck-master/defaults/src/main/resources/default.properties but is not exposed in the gui and is defaulted to true.

Desktop (please complete the following information):

  • OS: macOS
  • Version 9.4.1

Additional context
Add any other context about the problem here.

Metadata

Metadata

Assignees

Labels

No labels
No labels

Fields

No fields configured for Enhancement.

Projects

No projects

Milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions