Data Admin Module - Login Issue after Full User Data import/update

Twice now, I've tried to login to my Gibbon account and been told I have wrong password (though I didn't change it).

After thinking about what might have triggered the problem, I realized that I had run the extension Data Admin Module's "Full User Data" import, but in doing so I had left the password column empty (blank). This seems to be what caused the problem. I had exported this table earlier (using the Data Admin module), and had incorrectly assumed it would retain the original password information. Apparently it does not, and so I locked myself out of the system. Luckily I was able to retrieve the old password from my backup database and get back into the system, but just in case someone else runs into a similar problem I figured I'd post the problem and solution.

The User Data export/import also seems to cause a secondary issue with the "All Roles" column, wherein the numeric field becomes a single numeric string, instead of a list of numbers ("2001" instead of "002, 001"). At least this is what happens to me when I open the exported file in LibreOffice. This can cause a second login problem, related to "insufficient privileges", if a user has more than one role.

So note to anyone using this module (which is great BTW!), be careful to make sure that all the data is correctly entered and in the correct format before importing.

Comments

  • @sandra are you OK to look into this one? Thanks!
  • edited October 2018
    Hi Deanna,

    I agree with your conclusion. Importing a blank password for all your users (including yourself) would certainly lock you out of the system. A couple ways to avoid this might be to exclude yourself from the import, re-import the passwords each time, or select "Skip this column" for the password import so it doesn't overwrite them. Passwords aren't stored as plain text, so they can only be exported as hashes from the database: if they exist in plain text, it'll only be pre-import in your own files.

    The problem with All Roles might be a spreadsheet formatting issue, lots of spreadsheet tools will attempt to convert numbers and dates when opening a CSV. So, it's seeing 3 digit numbers separated by a comma, (002,001) and converting them to a whole number (2001). Depending on the software, there should be an option to turn off this kind of formatting on import. I know Excel loves to randomly turn student homerooms into dates, so it's definitely something to keep an eye on when working with CSV files, and try to use Plain Text for all the column formats.

    Hope this helps!
    admin
Sign In or Register to comment.