bug-hurd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Reauthentication EINTR bug


From: Carl Fredrik Hammar
Subject: Re: Reauthentication EINTR bug
Date: Mon, 5 Jul 2010 10:14:17 +0200
User-agent: Mutt/1.5.20 (2009-06-14)

Hi,

On Sun, Jul 04, 2010 at 12:34:19PM -0700, Roland McGrath wrote:
> The auth server, like any server handling interrupt_operation, is
> responsible for making sure that the operation either is entirely
> interrupted cleanly, or completes normally.  For the auth handshake,
> interrupting cleanly means that both sides of the transacation get
> interrupted.  In the scenario you describe, the rendezvous has already
> happened--it must have, for one side to be getting a successful RPC reply
> from the auth server.  So the auth server should not be letting the other
> side's RPC fail for interruption, it should complete.

This would be ideal, but my current understanding is that
interrupt_operation() is mostly advisory and that the operation will
restart or return EINTR regardless of what the server does, and if this
is true then it is impossible to complete the RPC.  However, my view of
the interruption code is not consistent enough for me to be certain so
I will investigate this possibility further.

Regards,
  Fredrik



reply via email to

[Prev in Thread] Current Thread [Next in Thread]