Windows Power Management was Re: [OpenAFS] OpenAFS 1.3.73

[Jeffrey Altman]

James:

Bugs are filed by sending e-mail to [EMAIL PROTECTED]
The bug report should attempt to include an unambiguous description of the problem and how it can be reproduced.

For example, the term "hanging" is extremely ambiguous.  It does
not describe the problem.  What does it mean?  What does it refer to?

Now in the case of a report I am very sure I do know what you are
referring to. The purpose of the PM code is to write all dirty buffers back to the file servers prior to the machine being suspended or
shutdown. Depending on the state of the network this may take time
or might not be possible at all.

If your "hanging" means you say you want to suspend and the computer
waits a minute before doing so then the PM code is currently attempting to do something and failing. If your "hanging" is something else
you should explain it in a more precise technical description.

The PM code in the Windows product was implemented at the wrong layer.
Instead of being added to the SMB client layer it should have been
added to the cache manager. The ways things are now the code will
not work properly on Terminal Server machines nor will it necessarily
do the right thing in other situations. In the OpenAFS Request Tracker there is an open ticket for this:

  15523: Windows Power Management fails to do the correct thing
    http://rt.central.org/rt/Ticket/Display.html?id=15523

Jeffrey Altman

James Burns wrote:

Jeffrey,
Do you have a check list of the steps a person should go through to
submit a bug? I'm still seeing issues probably related to power
management on my notebook in 1.3.73, though quite significantly less
than any previous release. It isn't crashing, but rather hanging, if
that makes a difference in the instructions. Thanks for your awesome
work on this software.

-James

smime.p7s

Description: S/MIME Cryptographic Signature