Re: Fwd: Re: gio synchronized unmount ?

[Alexander Larsson <alexl redhat com>]

On Sun, 2010-03-21 at 17:42 +0100, François Jaouen wrote:
> The I've tried to register to the signal mount-removed emitted by
> volume
> monitor but the result is weird ; sometime the signal handler is
> called
> before unmount_done_cb, sometimes it is the opposite and whatever the
> order is, the MTP device is not always available ?!
> 
> The only solution I've found is to add some kind of delayed retries to
> open the MTP device but I find this solution not very smart and I am
> surprised that there is not a more deterministic mean to be sure that
> the unmount operation is completely done.

In general the gvfs unmount system does ensure that we've sent the
unmount registering to the gvfs daemon before returning success on the
unmount operation. 

So, I would say that this is a bug in the gphoto backend. By the time
the unmount returns the operations *should* be completely done. However,
I don't really know how the gphoto stuff works, so i'm not sure why this
is not the case.


-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 Alexander Larsson                                            Red Hat, Inc 
       alexl redhat com            alexander larsson gmail com 
He's a one-legged Jewish hairdresser on the wrong side of the law. She's a 
cynical tomboy doctor who hides her beauty behind a pair of thick-framed 
spectacles. They fight crime!