manifoldcf-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Karl Wright <>
Subject Re: Disabling login on ManifoldCF 1.3
Date Tue, 06 Aug 2013 08:31:56 GMT
Hi Matt,

Unfortunately, this is not so simple.  The reason for the login is twofold:
Access security, but also because the ManifoldCF UI now keeps sensitive
information (existing passwords) in session variables.  See
CONNECTORS-737.  So, when a session expires, all bets are off and the user
is going to have to get redirected to the login page, so they know that
their session is lost.  Therefore, the basic UI flow must now be what it is.

If you can think of clever way to disable login without changing the UI
flow, please open a ticket and request an enhancement, including your
idea.  It would have to be something pretty lame - perhaps instead of
requesting credentials, there's just a link or button you can click that
says "Proceed to UI".


On Tue, Aug 6, 2013 at 4:10 AM, Matthew Sheppard

> I'd like to disable the new login for the ManifoldCF interface - Is this
> possible, and if so how?
> --- Background ----
> I currently have a copy of ManifoldCF included along with another
> application, and our setup arranges for our app server to authenticate
> users before they reach the ManifoldCF interface (mcf-combined-service.war).
> Since upgrading to 1.3 I see an additional login page (login.jsp) which
> appears after our login process completes. It looks like this new login
> page uses a username and password set via the
> and org.apache.manifoldcf.login.password
> properties (
> ).
> Since the double login and different username/password setup is likely to
> be confusing to our users, and since the new ManifoldCF system does not
> seem to allow us to use the same credentials as are used for the rest of
> our system easily, I'd like to turn off the ManifoldCF login leaving only
> our app server one.
> Looking briefly around I can't see any setting to disable the new login,
> but hopefully I'm missing something.
> ---- ----
> Many thanks for any assistance anyone can provide,
> Matt Sheppard

View raw message