Hi Nandana,
I will check on that when I get some time, too.

Thanks,
Dobri

On Mon, Sep 14, 2009 at 2:40 PM, Nandana Mihindukulasooriya <
[email protected]> wrote:

> Hi Dobri,
>       In the isSecHeaderRequired method, there is a logic for checking
> supporting tokens. That should cover Username Tokens and any kind of
> supporting tokens. But in your policy, your username token is not wrapped
> in
> any kind of supporting token policy assertion. So my guess is this is
> causing the issue. I will dig in this more, if I get some time. Can you
> please check on that.
>
> regards,
> Nandana
>
> On Mon, Sep 14, 2009 at 1:09 PM, Dobri Kitipov <
> [email protected]
> > wrote:
>
> > Hi all,
> > I have noticed that if I have the following security policy:
> >
> >    see the attachment.
> >
> > , or Asymmetric binding + timestamp and UsernameToken. IMHO it is a valid
> > case to have the above mentioned policy without the timestamp but only
> with
> > UsernameToken.
> >
> > The problem is that when RampartEngine.process(MessageContext) is invoked
> > then in turn it invokes
> RampartUtil.isSecHeaderRequired(RampartPolicyData,
> > boolean, boolean)
> >
> > here we check if the security header is required. And if we have check
> for
> > the timestamp:
> >
> >         // Checking for time stamp
> >         if ( rpd.isIncludeTimestamp() ) {
> >             return true;
> >         }
> >
> > We do not have check for the Username token. IMHO we need the same check
> > for the Username , too.
> > Please, provide me with your comments. I am a little bit confused why
> such
> > check is not available?
> >
> > If I am right I can commit the needed changes.
> >
> > Thanks,
> > Dobri
> >
>

Reply via email to