Re: Is Heimdal thread safe?

[Alexey Melnikov <Alexey.Melnikov@MessagingDirect.com>]

On 06 Sep 1999 05:39:12 +0200 Assar Westerlund <assar@sics.se> wrote:

> joda@pdc.kth.se (Johan Danielsson) writes:
> > 
> > It's a very simple function that just compares two names, and tells
> > whether they are equal. But if you need more that just the names
> > themselves to do this, you're smoked. In the Heimdal case you need a
> > krb5_context, and the only way to inject a krb5_context in this API is
> > via a global variable.
> 
> Or you could do it with a thread-global variable, but that's also a
> pain.
> 
> For now I would suggest having a lock around the calls to the
> gss-api.  Would that cause too much contention in your application?

Hopefully not, but I can't tell anything until I try.

Do you know whether MIT implementation is thread safe?

Independent question: Heimdal and MIT were tested against each other, 
right? Do they interoperate?
---------------
Alexey Melnikov
Software Developer    phone: (780) 424 4922 x 357
MessagingDirect Ltd.  fax  : (780) 424 4925

mailto:mel@messagingdirect.com
http://www.MessagingDirect.com

Come see us at these Trade Shows:
ICE '99 Oct 19-21/99 in San Francisco   Booth 859
ISPcon  Oct 26-28/99 in San Jose        Booth 1845 

* This e-mail message was sent with Execmail V5.0 *