Announcement Regarding Support

Hello everyone, Charles here. Please read my post on changes to Media Cloud support and updates regarding Media Cloud in general.


Files are already on S3, how to get MediaCloud to recognize them?

  • @jasondiff

    I looked into adding a file name filter, but that's not possible with S3 or S3 compatible cloud storage providers.

  • Ah, I thought you were talking about adding a page/limit/offset ability. But I'll try the path filtering to do it in batches.

    I just heard from WPEngine that the current status is [2715 of 458352] which seems awful slow (it's been over 24 hours) but I'm inclined to just let it keep going.

  • @jasondiff

    478K! I'll be super honest, the import functionality is not very optimized. Hopefully now that it's started it'll go quick.

    LMK if there are any import issues.

  • So I'm getting the import to work, albeit slowly, but I am able to use the wordpress backend and select folders to import subsets of images at a time.

    Here's the issue I've discovered.


    The import is not connecting the cloud image to the existing entry in the media library, but instead adding a new entry. So the images that had already been put in the media library and attached to existing posts still won't be served from the cloud.

    Is there a way to fix this other than doing a full upload of the media library to the cloud (which is exactly what I was trying to avoid since all the images are on S3 already)?

    [Side note - I know why the webp version is there, I neglected to delete those from S3 before running the import. I'll fix that for future imports.]

  • @jong

    I know you've been busy with WP5.3 issues but just wondering if you have any thoughts on this?
    I'm going to upgrade to the newest version of Media Cloud and give it another try.

  • Can confirm that MediaCloud 3.3.0 exhibits the same behavior.

  • @jasondiff

    The import process requires Media Cloud to guess what existing media in your library matches to whatever it is you are trying to import.

    It tries to match based on the guid of your attachment posts. The guid for most attachements is it's url on your server or, if uploaded with Media Cloud or Offload Media, would be the URL on cloud storage at the time it was uploaded.

    So Media Cloud tests to see if the imported file, if it were served from your local server, matches the GUID of any attachment. If it does, than the imported file is tied to that attachment.

    If it can find a GUID for that test, then it uses the URL of the thing you are importing. If that URL matches a GUID than it uses that to tie the imported file to the attachment.

    There's no real other test that Media Cloud can do to match an imported file to a file in the media library. Going by filename, which seems obvious, doesn't work because common filenames would always match with each other.

  • @jasondiff

    But now that I think of it there may be another way to do the match up too. Let me add it to the pile of stuff I have to get through, maybe by the weekend I'll be able to tackle it.

  • That would be great! Keep me posted!

  • @jasondiff

    Fixed in 3.3.2.

Log in to reply