In message <[EMAIL PROTECTED]>,David Howells writes:
>> >    Session Keyring
>> >           -3 --alswrv      0     0  keyring: _uid_ses.0
>> >            2 --alswrv      0     0   \_ keyring: _uid.0
>> >     29391168 ----s--v      0     0   \_ afs_pag: _pag
>> >
>
>What I'm pointing out in the above key ring dump is the fact that the _pag key
>is attached to the _uid_ses.0 keyring (the root user default session keyring).

ah right i see now.  sorry, i have come in a bit late on this.  i would
guess this an oversight on my part.  the afs install_session_keyring()
isnt checked for failure.  so if it does fail and there happens to be
a session keyring already present it will get the pag.  this is not the
intended behavior.  of course pagsh() cant tell you it failed either.
still afs shouldnt set a pag if it didnt manage to create a session keyring.

>> the afs pag is always uid = 0 so that users cant modify the key and discern
>> its contents.
>
>Can you achieve that by just not providing read and update ops for the key?

i dont have read or update ops now.  i dont think this would be sufficient
since the afs_pag key type still has to have an instantiate op which
the user could call.  i dont want users creating session keyrings and
arbitrary pags trying to join existing pags.  particulary since pags are
given out in a serial fashion.  (someone should fix this).
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info

Reply via email to