<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">On 23/08/16 15:04, Bill Burke wrote:<br>
    </div>
    <blockquote
      cite="mid:f08b9c3c-6c39-037e-fa7b-c32cf5db9390@redhat.com"
      type="cite">
      <meta content="text/html; charset=windows-1252"
        http-equiv="Content-Type">
      <p><br>
      </p>
      <br>
      <div class="moz-cite-prefix">On 8/23/16 3:39 AM, Marek Posolda
        wrote:<br>
      </div>
      <blockquote cite="mid:57BBFDCB.4080705@redhat.com" type="cite">
        <meta content="text/html; charset=windows-1252"
          http-equiv="Content-Type">
        <div class="moz-cite-prefix">On 19/08/16 15:52, Bill Burke
          wrote:<br>
        </div>
        <blockquote
          cite="mid:1d491ec6-14d8-f1f8-3dbe-ad6b5c482f66@redhat.com"
          type="cite">
          <meta content="text/html; charset=windows-1252"
            http-equiv="Content-Type">
          <p><br>
          </p>
          <br>
          <div class="moz-cite-prefix">On 8/19/16 2:37 AM, Stian
            Thorgersen wrote:<br>
          </div>
          <blockquote
cite="mid:CAJgngAcrFJurnNYFB3Vi1dWmkVrWWWjtVVTMgp9C+VFtgxFFMQ@mail.gmail.com"
            type="cite">
            <div dir="ltr"><br>
              <div class="gmail_extra"><br>
                <div class="gmail_quote">On 18 August 2016 at 20:30,
                  Bill Burke <span dir="ltr">&lt;<a
                      moz-do-not-send="true"
                      href="mailto:bburke@redhat.com" target="_blank"><a class="moz-txt-link-abbreviated" href="mailto:bburke@redhat.com">bburke@redhat.com</a></a>&gt;</span>
                  wrote:<br>
                  <blockquote class="gmail_quote" style="margin:0 0 0
                    .8ex;border-left:1px #ccc solid;padding-left:1ex"><span
                      class=""><br>
                      On 8/18/16 4:59 AM, Stian Thorgersen wrote:<br>
                      &gt; Bill,<br>
                      &gt;<br>
                      &gt; Are you planing to have an option to allow
                      import of users with the<br>
                      &gt; new user federation SPI? I'm not convinced we
                      should completely remove<br>
                      &gt; this option.<br>
                      &gt;<br>
                      <br>
                    </span>The only callback that does not exist in the
                    new SPI is<br>
                    validateAndProxy().  With the current federation
                    SPI, the developer<br>
                    implements everything themselves for import.  There
                    are no<br>
                    synchronization APIs/SPIs either.<br>
                  </blockquote>
                  <div><br>
                  </div>
                  <div>Sounds like we're removing built-in features
                    around synchronization just to make the user have to
                    do everything themselves.</div>
                </div>
              </div>
            </div>
          </blockquote>
          I think you misinterpreted me,  The old User Federation SPI
          forces the developer to write all the import code themselves. 
          The old User Federation SPI does not have any synchronization
          callbacks, methods or interfaces other than
          validateAndProxy(), the logic of which the user has to write
          themselves too.<br>
          <br>
          <br>
          <blockquote
cite="mid:CAJgngAcrFJurnNYFB3Vi1dWmkVrWWWjtVVTMgp9C+VFtgxFFMQ@mail.gmail.com"
            type="cite">
            <div dir="ltr">
              <div class="gmail_extra">
                <div class="gmail_quote">
                  <div> </div>
                  <blockquote class="gmail_quote" style="margin:0 0 0
                    .8ex;border-left:1px #ccc solid;padding-left:1ex"> <span
                      class="">&gt; Some use-cases I could imagine:<br>
                      &gt;<br>
                      &gt; * Allow users to authenticate even if LDAP
                      server is down<br>
                    </span>Our current LDAP provider will not work if
                    LDAP is down, even with the<br>
                    import :)<br>
                  </blockquote>
                  <div><br>
                  </div>
                  <div>Yes, I know. However, the fact that we don't
                    currently support it doesn't mean we shouldn't in
                    the future.</div>
                </div>
              </div>
            </div>
          </blockquote>
          If the user can only be authenticated via LDAP, an offline
          mode is not possible.  In other words, if LDAP does not expose
          the password of a user (so it can be imported), then offline
          mode is not possible.  It would only be possible if the user
          has logged in at least once, then the validated password could
          be imported.<br>
        </blockquote>
        <blockquote
          cite="mid:1d491ec6-14d8-f1f8-3dbe-ad6b5c482f66@redhat.com"
          type="cite"> <br>
          So, do you still think we should support import/offline mode
          given all this?<br>
        </blockquote>
        From some recent discussions I saw, it seems that quite many
        people are interested in the "import-and-forget" mode. So they
        need to import user from their old legacy store (3rd party
        storage or LDAP) but once user is fully in Keycloak DB, they
        want to completely forget about the 3rd party storage and do all
        operations around this user against Keycloak DB.<br>
        <br>
        The credentials/password validation seems to be the most
        complicated part around this as you pointed, as the password
        needs to be first successfully validated against 3rdparty
        storage or LDAP . Then once password is successfully validated
        and updated to Keycloak DB, user can be "forgotten" and unlinked
        from the federationProvider. I hope the new SPI has a way to
        deal with this usecase? Or at least have a hook, so the people
        can easily unlink the user by themselves whenever they want.<br>
        <br>
      </blockquote>
      As I said  before, the current SPI does not have any support for
      import.  It also does not have any SPIs around synchronization or
      any synchronization buttons in the admin console.  It is up to the
      developer to write the code to import the user.  And our current
      LDAP implmementation is not "import and forget", you already
      mentioned password validation, but there is also validateAndProxy
      which is called every time the user is accessed and which hits
      LDAP every time.  There's also no way to unlink the user. <br>
    </blockquote>
    Not right now, but it seems that many people consider the
    "import-and-forget" as important usecase? You just want to import
    the users from 3rd party storage or LDAP, but you need to do in
    multiple steps and "wait" until password is successfully validated
    for the first time.<br>
    <br>
    As an example this blogpost from Scott Rossillo
    <a class="moz-txt-link-freetext" href="https://tech.smartling.com/migrate-to-keycloak-with-zero-downtime-8dcab9e7cb2c#.1e8sy1o8n">https://tech.smartling.com/migrate-to-keycloak-with-zero-downtime-8dcab9e7cb2c#.1e8sy1o8n</a>,
    which AFAIK seemed to have some positive feedback from more
    community users.<br>
    <br>
    I don't know how deeply to go with directly supporting it at SPI
    level. However IMO it will be good to have at least same level like
    the current UserFederation SPI. So at least at some point (ie. after
    successful password validation), the people can manually unlink the
    3rd party provider from the user and migrate all the data to
    Keycloak DB and then use it just from Keycloak DB.<br>
    <br>
    Marek<br>
    <blockquote
      cite="mid:f08b9c3c-6c39-037e-fa7b-c32cf5db9390@redhat.com"
      type="cite"> <br>
      So, #1, LDAP users are getting no real advantage to importing into
      our database with tremendous overhead<br>
      #2, I do not think we have many people implementing their own
      custom providers.  Why?  We've had few questions on something that
      is really really hard to do (see the whole deployer discussion).<br>
      <br>
      I just think this whole import thing is a bad idea.<br>
      <br>
      Bill<br>
    </blockquote>
    <br>
  </body>
</html>