Support

Good evening,

Just wanted to report this because I'm not sure if it may be a situation that is overlooked currently.

I apparently have just a single file that somebody uploaded last year on our live Linux-based web server, where colons in the filename aren't an issue ("Monthly Attendance Verification 8:28:19.pdf") and after generating a backup and trying to restore it locally in my Windows XAMPP setup at home, it failed due to the filename above.

I just wanted to throw out the suggestion, if there could be a potential option in Akeeba to skip one or more files with colons in the name specifically when restoring from a Windows machine since that would result in a known issue due to Windows not allowing colons in the filename.

Additionally, I'm pretty sure that my DOCman install (due to it being used from DOCman 1.5 up until the current version) still has some files in it that are similar in name that would result in overlap on Windows (e.g. "Filename1.jpg" and "filename1.jpg" would overlap on Windows due to the lack of case sensitivity for the filenames). When copying the files from SFTP over to a Windows machine, WinSCP prompts if I want to override the files in this case and I can then choose the newer of the two files. I'm not sure how Akeeba handles this particular scenario behind the scenes, but thought I'd mention it as well.

Of course, this isn't typically an issue with backups/restorations generally occurring on Linux (or Unix) based OSes, but could cause some issues for less techie individuals.

In my case, I'm going to see if the file in question is actively being used anywhere and if not, I'm going to remove it from DOCman and create a new Akeeba Backup which will hopefully be able to be restored successfully.

One additional note...there was also a file named "(1)." which ends in a period, which is another type of filename that's not allowed on Windows so this was another spot where the restoration process broke down for me with the next backup I made. (This Stack Overflow article has some useful info on the allowed filename rules: https://stackoverflow.com/questions/1976007/what-characters-are-forbidden-in-windows-and-linux-directory-names)

I thought the particular file was a PDF that had been uploaded incorrectly, but not quite sure what the filetype actually was (viewing the file in a text editor seemed to indicate it might be related to Adobe XMP filetype, but I'm not familiar with that one, so I'm guessing it may have come in as part of an upload of a fancier PDF file that a staff member created, but not 100% sure). I ended up deleting the file from the server since it didn't seem to be related to any existing DOCman documents, and am re-running the backup a 3rd time (hopefully this time the restoration goes through successfully).

Makes sense Davide, sounds like that would be the option to enable in this situation, good to know for next time ;-). Thank you!

This ticket has been automatically closed. All tickets which have been inactive for a long time are automatically closed. If you believe that this ticket was closed in error, please contact us.