Windows CCAPI design sketch

[Jeffrey Altman]

Alexandra Ellwood wrote:
>
> On Jan 25, 2008, at 10:45 AM, Jeffrey Altman wrote:
>
>> Kevin Koch wrote:
>>> I've updated the Windows CCAPI design sketch, hopefully answering 
>>> previous
>>> questions.
>>> The document has moved again and is now at
>>> http://k5wiki.kerberos.org/wiki/Projects/Windows_CCAPI.
>>>
>>> For those who are interested, the source code has been checked in to
>>> .../krb5/src/ccapi/*/win.
>>>
>>> Comments are welcome.
>>>
>>> Thanks.
>>>
>>>
>> Kevin:
>>
>> Could you please add sections describing:
>> * the platforms that are being supported
>> * the development tools that are being supported
>
> Don't know anything about these Windows issues.
I wasn't expecting you to be the one responsible for answering them.  
>
>>
>> * a summary of the differences in functionality between the currently 
>> deployed CCAPI and this one.  In particular, what are the known 
>> compatibility issues that application developers are going to have to 
>> deal with when the transition occurs.
>>
>
>
> Cross-platform changes:
>
>
> CCAPI v2 is deprecated.  All CCAPI v2 functions return errors when 
> called.
This decision is going to cause problems for all currently deployed 
applications which support the use of multiple credential caches.  These 
applications use the CCAPIv2 to enumerate the available credential 
caches in order to determine which caches are available.   There has 
been no other interface for them to use.  By failing to implement 
CCAPIv2 on top of the new implementation there will be no transition 
mechanism for organizations to use when upgrading to KFW 4.0 and CCAPIv7.

>
> Locking functions (cc_context_lock, cc_context_unlock, cc_ccache_lock 
> and cc_ccache_unlock) now work.  Locks are advisory so existing 
> callers should see no change in behavior if they were not using the 
> locking functions.  Obviously if code was using the locking functions 
> previously (in the previous implementation they just functioned as 
> NOPs) then the caller might see different behavior switching to the 
> new CCAPI.
>
> Two new functions: cc_context_wait_for_change() and 
> cc_ccache_wait_for_change() which block until the next change to the 
> cache collection or ccache respectively.  These function allow 
> multithreaded applications to avoid polling when trying to update on 
> changes.  They are intended to replace the "change time" functions.
>
> In order to allow applications to test for these new features the API 
> version number has been bumped to CCAPI version 7.
>
> Any other difference in the behavior of the new and old CCAPI 
> implementations is a bug and should be filed in our bug tracking system.
These are changes to the CCAPI for versions that were never deployed on 
Windows.  Therefore, I do not consider them applicable to the discussion 
of the Windows implementation.   The questions that need to be answered 
are those that application developers and system administrators are 
going to use to determine how and when KFW 4.0 can be deployed


Jeffrey Altman

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3355 bytes
Desc: S/MIME Cryptographic Signature
Url : http://mailman.mit.edu/pipermail/kfwdev/attachments/20080130/b982830c/attachment.bin