Issue: MediaWiki Images not imported

VaultWiki passes the URLs to vBulletin's attachment system, vBulletin uploads them as if you tried to upload via URL (rather than your computer), vBulletin stores them like any other attachments.

If your URL is correct, are the images publicly accessible? I have seen some users put a password protection on the MediaWiki directory because they intended to close it. If you do this, vBulletin can't access the images to import.

Are the images very large in file size or dimensions? If the images are bigger than your vBulletin attachment settings allow, vBulletin won't try to import them.

Does the server the MediaWiki installation is on allow HTTP requests using the HEAD method (as opposed to GET or POST)? vBulletin fetches metadata about the image before uploading, via HEAD request. If the server doesn't allow HEAD requests, vBulletin won't be able get this data, and won't try to import the image.

Are the source images corrupt or compressed? I have seen users unable to import images because they were unaware the source hard drive had failed and corrupted many of the images.

Thank you for the detailed response. I'll look into each of these potential issues.

No success on any of these. Debugging mediawiki.php I can see that in function forge_article(&$page) it never attempts to import because "if ($page['thread_id']" always fails the test.

Suggestions?

What other modifications do you have installed? Obviously something is preventing the thread from being created if the threadid is always missing.

I have vbGarage installed though I suspect that is not the issue.

On the very first import attempt the path to the MediaWiki images was wrong but it imported the articles. I corrected the path and I have tried multiple times to import, and each time all it does is re-import all the articles over, but no images.

I do not understand the wording on the Import screen where it says "A previous import was detected." It is unclear to me what is the difference between answering yes or no. What I want to do is have it completely rewrite the previous import but that doesn't seem to be an option. Could this be what is going wrong?

I am doing this work on a "sandbox" server because I don't want to try it on my production system until I understand how to make it work correctly.

You will need to purge or cleanup the image forum that was affected by the incorrect path before you can import again.

On "A previous import was detected":

Yes, ignore duplicate IDs - This will result in getting 2 copies of articles. Rather than adding missing revisions to old articles, it creates a new copy of the article with the imported revisions. You need to sift through the duplicates manually, and this can only be done via the AdminCP, since duplicates share the same URL but have different content. That's one of the functions of the Wiki URL Manager. This is mainly useful if you're importing articles from multiple sources that might have name collisions.

No, respect duplicate IDs - If an article with the same name already exists, it won't be imported. New edits to the article will still be imported, since we can match them to the article ID.

In both cases, choosing Yes or No, it doesn't remove already imported content - in this case images that were imported with the wrong path. You will be stuck with them until you cleanup the image forum and import that forum again.

Deleting all the incorrectly imported images and running the import again worked. Thank you.