[cups.development] Re: [cups.commit] [CUPS] r5589 - branches/kerberos/scheduler

[Daniel Henninger]

*grin*  Is someone working on Kerberos auth that I should be aware  
of?  There's certainly little point in me working on it if someone  
else has already dove into it.

Daniel

On May 25, 2006, at 2:06 PM, Michael Sweet wrote:

> cups-dev at easysw.com wrote:
>> ...
>> --- branches/kerberos/scheduler/client.c	2006-05-24 19:37:25 UTC  
>> (rev 5588)
>> +++ branches/kerberos/scheduler/client.c	2006-05-24 21:47:55 UTC  
>> (rev 5589)
>> @@ -1966,10 +1966,13 @@
>>    /*
>>    * To work around bugs in some proxies, don't use Keep-Alive for  
>> some
>> -  * error messages...
>> +  * error messages... +  *
>> +  * Kerberos authentication doesn't work without Keep-Alive, so  
>> +  * never disable it in that case.
>>    */
>>  -  if (code >= HTTP_BAD_REQUEST)
>> +  if (code >= HTTP_BAD_REQUEST && con->http.auth_type ==  
>> AUTH_KERBEROS)
>>      con->http.keep_alive = HTTP_KEEPALIVE_OFF;
> > ...
>
> Isn't this logic wrong?  Based on the comment, don't you want:
>
>     if (code >= HTTP_BAD_REQUEST && con->http.auth_type !=  
> AUTH_KERBEROS)
>         con->http.keep_alive = HTTP_KEEPALIVE_OFF;
>
> -- 
> ______________________________________________________________________
> Michael Sweet, Easy Software Products           mike at easysw dot com
> Internet Printing and Publishing Software        http://www.easysw.com
>
> _______________________________________________
> cups-dev mailing list
> cups-dev at easysw.com
> http://lists.easysw.com/mailman/listinfo/cups-dev