Feeds 7.x-2.0-beta3 is planned to be released on November 24, 2016. See #2826400: Plan for Feeds 7.x-2.0-beta3 release.

For Feeds 7.x-2.0-beta4 somewhat bigger changes are planned in order to fix major bugs.

  1. Cron issues
    There are a lot of reports about imports that do not work during cron. We hope to fix at least some of these by running the imports during cron as either user 1 or the user who created the feed node.
    #2541944: Taxonomy mapping does not work with Cron
    #1995728: [META] Cron import not working on 7.x-2.0-alpha8 and later
  2. Validation
    Feeds will be stricter on validating imports to minimize SQL errors on imports. Furthermore, importers itself will be validated too. At first only for if the configured bundle and language on the processor exist. Feeds will also no longer allow attaching two importers to the same content type. Some settings will get locked when there is already imported data.
    #2379631: field_attach_validate() must be called before programmatic entity saves
    #2474589: Changing processor wipes mappings with no warning
    #2640108: Display an error message when attaching an importer to a content type that is already in use by an other importer
    #2320781: Validate feed importer configuration: check for invalid bundle and invalid language
  3. HTTP Fetcher issues
    There are two big issues with the HTTP fetcher. They are both related to imports that need multiple cron runs to complete. One is that a source is entirely stored in the database. Second is that a source to import from is stored as a temporary file, which is removed after 6 hours. If the import hasn't been completed by then, the import will be aborted and the importer will be stuck in a locked state.
    #2032333: [Meta] Store remote files in the file system instead of in the database
    #1029102: Importing Large CSV document (downloaded and processed in the background)


MegaChriz created an issue.