Hi all, Is there any solution to search for corrupted content already saved? Using DQL or Custom Java?
We create content using custom Java and REST but some documents has corrupted content - It canot be dumped or saved.
Thx all, Dan.
Can you be more specific? What exactly is corrupted? Are you saying that the file size is different from the source file you imported? Is it for all file formats or just particular formats? Do you have a virus scanner running on the Content Server machine? Is the file size different from dmr_content.full_content_size (or dm_sysobject.r_full_content_size)?
When you upload a file to Documentum, the server preserves the file in binary format (ie it doesn't try to read the contents). Therefore, the server has no way to "know" whether the binary file is corrupt or not. Do you have CTS installed in your environment? If CTS cannot generate rendition of file (assuming the file type is supported type - eg Word), this may be a sign that the file is corrupt. But, you can also get plenty of false positives - where rendition generation failure is caused by something else. You can query server for which files do not have renditions or alternatively you can query for the rendition queue to see which documents are failing to generate renditions.
There are CTS servers on the environment but uploaded files are already pdf.
There might be some libraries on the internet that can check for PDF file validity. You will need to write your own custom DFC utility to query dmr_content objects for pdf format, call this library, and then log whether its valid or not.
