Yes/no... It's an "anonymous user", but it's still a connection that you
established with your password.

i.e. map to //host/username, then to //host/public, where public is a
guest accessible share.

Also, it doesn't appear this is completely reliable behavior, as I can
reproduce it sometimes, but not others. (Not sure 100% what causes the
setgroups() to be issued, it doesn't get issued in all cases.)

I just figured that this behavior is different than all the other
initgroups/setgroups behaviors. They both maintain the pag for all cases
other than the empty-group-list case. If it passes a list of groups -
i.e. if 'nobody' had some aux groups, it would maintain the pag. 

-- Nathan

------------------------------------------------------------
Nathan Neulinger                       EMail:  [EMAIL PROTECTED]
University of Missouri - Rolla         Phone: (573) 341-4841
Computing Services                       Fax: (573) 341-4216


> -----Original Message-----
> From: chas williams [mailto:[EMAIL PROTECTED]] 
> Sent: Thursday, May 23, 2002 11:10 AM
> To: Neulinger, Nathan
> Cc: [EMAIL PROTECTED]
> Subject: Re: [OpenAFS-devel] interesting issue with setgroups 
> wrapper... 
> 
> 
> In message 
> <[EMAIL PROTECTED]>,"Neulinge
> r, Nathan" writes:
> >Not between multiple users, between the user, and anonymous
> >non-user-based connections on the same smb session. (All that screwy
> >stuff windows does with multiple samba connections from the same PC.)
> >The assumption is that a single smbd has a single afs authentication
> >context is still kept.
> 
> if i read this correctly then the user's smbd is switching between
> the user and the anonymous user.  you wouldnt want the anonymous user
> to be able to user your pag so smbd is dropping all the groups when 
> switching from the user to the anonymous user.  it probably 
> restores the
> groups with initgroups() instead of saving the groups with getgroups()
> and restoring with setgroups()?
> 
> 
_______________________________________________
OpenAFS-devel mailing list
[EMAIL PROTECTED]
https://lists.openafs.org/mailman/listinfo/openafs-devel

Reply via email to