Skip to content

Security issue: File URLs are easy to guess #9

@rimas-kudelis

Description

@rimas-kudelis

With the file urls like http://SERVER.TLD/file?id=123, it's very easy to see any file on the server: you just start with ?id=1, and keep incrementing the number for as long as you like. To me, it seems like quite a huge security issue.

To work this around, I suggest to use a different identifier for file queries. For example, storing file id as GUID, which is not incremental, might work well.

Also, I think it would be nice to be able to link to the file directly, if it's uploaded to a public place, although if the above were implemented, that would be enough.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions