• Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Minor Minor
    • _unsorted
    • None
    • Jenkins 2.46.1
      Email Extension 2.57.2

      This may be releated to #43178.

      I have another Jenkins instance that uses the Jenkins internal database. I just upgraded this instance. I have 2 entries on the people page:

      Userid: "Jon Schewe", full name: "Jon Schewe", email "jpschewe@xxx", delete button is active, no password is set. Last commit activity is on my last job.

      Userid: "jpschewe", full name: "Jon Schewe", email "jpschewe@xxx", delete button is NOT active, password is set. Since the password is set, presumably this is who I'm logging in as. There is no last commit activity on this user.

      I committed changes as "jpschewe@xxx" through GitHub. The github hook fired and caused my build to run. At the end of the build:
      19:33:37 Email was triggered for: Always*19:33:37* Sending email for trigger: Always*19:33:37* Not sending mail to unregistered user jpschewe@xxx*19:33:37* An attempt to send an e-mail to empty list of recipients, ignored.
       

      This is using the Jenkins user database, so I would expect this is the case that was tested for, as opposed to my other instance that uses the unix database. This security fix doesn't appear to be working as expected. I understand the concept of not sending emails outside an organization, however the fix that I had before of using a presend script to filter email addresses has worked rather nicely. I'm not sure why Jenkins has 2 users for me in the people list or how this effects this sending of email, but I suspect they are related.

       

          [JENKINS-43759] Not sending email to registered users

          David van Laatum added a comment - danielbeck

          Daniel Beck added a comment -

          jpschewe To clarify, this isn't the "mailing list" issue you described in JENKINS-43178?

          davidvanlaatum I thought I explained somewhere that these users are considered different (Jon describes this in a way I suspect he read my explanation, but it's not in JENKINS-43178). What could be considered here is to search for a legitimate user with the same email address, but that would be an RFE. From my POV this works as intended, even if annoying. If the Git committer's username were 'jpschewe', things would work.

          Daniel Beck added a comment - jpschewe To clarify, this isn't the "mailing list" issue you described in JENKINS-43178 ? davidvanlaatum I thought I explained somewhere that these users are considered different (Jon describes this in a way I suspect he read my explanation, but it's not in JENKINS-43178 ). What could be considered here is to search for a legitimate user with the same email address, but that would be an RFE. From my POV this works as intended, even if annoying. If the Git committer's username were 'jpschewe', things would work.

          jpschewe added a comment -

          danielbeck This is not the "mailing list" issue. This is a different Jenkins instance using the Jenkins database for access control.

          jpschewe added a comment - danielbeck This is not the "mailing list" issue. This is a different Jenkins instance using the Jenkins database for access control.

          Daniel Beck added a comment -

          Looks superficially like a duplicate of JENKINS-43268 but will need further looking into.

          Daniel Beck added a comment - Looks superficially like a duplicate of JENKINS-43268 but will need further looking into.

          Jesse Glick added a comment -

          Very likely an issue in git not email-ext, like JENKINS-9016 perhaps.

          Jesse Glick added a comment - Very likely an issue in  git not email-ext , like  JENKINS-9016 perhaps.

            Unassigned Unassigned
            jpschewe jpschewe
            Votes:
            4 Vote for this issue
            Watchers:
            7 Start watching this issue

              Created:
              Updated: