Delete someone’s account

WRT handles these requests - forward any such requests to them via email.

The WCA has processes for anonymizing Results data – i.e. “soft deleting” a WCA ID. We do not however comply with any requests to delete accounts in the stricter sense, i.e. username/password combinations for people to log in with

Remove an avatar

  1. Delete the avatar from the S3 bucket
    1. Bucket can be accessed with this link: https://s3.console.aws.amazon.com/s3/buckets/wca-avatar?region=us-west-2&prefix=uploads/user/avatar/&showversions=false
      1. Alternatively: Naviate to AWS Console -> S3 -> Buckets -> wca-avatar -> uploads -> user -> avatar
    2. Navigate to the user’s WCA ID
    3. Select all items you want to delete (delete the avatar in question and the thumbnail for it)
    4. Type the required text in the confirmation page
  2. Invalidate the avatar from the avatars Cloudfront distribution
    1. Search “Cloudfront” in AWS
    2. Select the distribution with “avatars” in the “Alternate destination” and/or “Origins” columns
    3. Go to “Invalidations” -> “Create Invalidation”
    4. Create an invalidation with the URL where the avatar lives e.g /uploads/user/avatar/2019POLA01/1569217830.jpg

AWS Certificate Renewal

AWS will send an email with the title “Action Required - Your certificate renewal”, which specifies the domain that needs a certificate renewed - for example, http://statistics-api.worldcubeassociation.org/

  1. Go to Certificate Manager in AWS console
  2. “List certificates”
  3. Click on the certificate with the relevant name in the “Domain name” column
  4. Click “Create records in Route 53”
  5. Confirm with “Create records”

Resetting a Password

We never reset a password on someone’s behalf - they should use the built-in password change feature.

Results Export/Developer Export has not run for a while

  1. check the cronjob_statistics database table for either DumpDeveloperDatabase or DumpPublicResultsDatabase
  2. Study the contents of the relevant db row - for example, last_error_message may help you understand why it didn’t run.
  3. For the job to run, it must have a run_end time - if you would like to rerun the job, you can
    1. Manually set the run_end time in the database
    2. Run the following command in the production console: bin/rake db:dump:development or bin/rake db:dump:public_results