no lower_user/lower_pass in pre2 ?
In previous versions of radiusd.conf there has been: lower_user = after lower_pass = after In pre2 this is missing and a grep -r lower_pass on the source tree does not show anything. Is this feature replaced my another one? Norbert Wegener
Norbert Wegener wrote:
In previous versions of radiusd.conf there has been:
lower_user = after lower_pass = after
In pre2 this is missing and a grep -r lower_pass on the source tree does not show anything. Is this feature replaced my another one?
No. I understand the utility of it, but it was a nighmare to support in the server. You can get *almost* the same functionality with unlang. What, specifically do you need? Why? Alan DeKok.
Alan DeKok wrote:
Norbert Wegener wrote:
In previous versions of radiusd.conf there has been:
lower_user = after lower_pass = after
In pre2 this is missing and a grep -r lower_pass on the source tree does not show anything. Is this feature replaced my another one?
No. I understand the utility of it, but it was a nighmare to support in the server.
You can get *almost* the same functionality with unlang.
What, specifically do you need? Why?
CAPS LOCK has always been a nice problem and lower_user as well as lower_pass solved that without any trouble. If this can be solved by unlang, fine. Norbert Wegener
Alan DeKok. - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
Norbert Wegener wrote:
CAPS LOCK has always been a nice problem and lower_user as well as lower_pass solved that without any trouble. If this can be solved by unlang, fine.
Maybe with "tr" functionality. It's not in 2.0, but it looks to be not hard to add. Running the request through the server twice is wrong. Alan DeKok.
On Fri, 2007-09-07 at 14:26 +0200, Alan DeKok wrote:
Norbert Wegener wrote:
In previous versions of radiusd.conf there has been:
lower_user = after lower_pass = after
In pre2 this is missing and a grep -r lower_pass on the source tree does not show anything. Is this feature replaced my another one?
No. I understand the utility of it, but it was a nighmare to support in the server.
You can get *almost* the same functionality with unlang.
What, specifically do you need? Why?
On thing I have used this for is with mac-address auth, to convert upper-case mac address formats to lower-case. A "tr" module in the server would be very useful.
Phil Mayers wrote:
On thing I have used this for is with mac-address auth, to convert upper-case mac address formats to lower-case.
It's better to just use lowercase for everything, rather than running an authentication request *twice* through the server. And the new configuration possibilities in 2.0 make it easy to emulate the "lower_password", etc. configurations.
A "tr" module in the server would be very useful.
Yes. Alan DeKok.
On Sun, Sep 09, 2007 at 06:26:37PM +0200, Alan DeKok wrote:
Phil Mayers wrote:
On thing I have used this for is with mac-address auth, to convert upper-case mac address formats to lower-case.
It's better to just use lowercase for everything, rather than running an authentication request *twice* through the server.
Well yes - but if the NAS sends it in upper-case...
And the new configuration possibilities in 2.0 make it easy to emulate the "lower_password", etc. configurations.
Indeed.
A "tr" module in the server would be very useful.
Yes.
I did look into this briefly; sadly the "tr" source code is pretty big and not exacltly modular :o)
Phil Mayers wrote:
It's better to just use lowercase for everything, rather than running an authentication request *twice* through the server.
Well yes - but if the NAS sends it in upper-case...
Then mash it to lowercase...
I did look into this briefly; sadly the "tr" source code is pretty big and not exacltly modular :o)
GNU tr? Don't look at any of *that* stuff. The BSD tr's will be much smaller and cleaner. And in any case, we would need very little of the full 'tr' functionality. Even a minimal version would be better than nothing. Alan DeKok.
participants (3)
-
Alan DeKok -
Norbert Wegener -
Phil Mayers