Re: sqlite3

[Eli Zaretskii]

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: emacs-devel@gnu.org
> Date: Fri, 17 Dec 2021 09:25:51 +0100
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > If Emacs 1 attempts to open and read the file while Emacs 2 still
> > writes to it, Windows will not let Emacs 1 open the file, and the
> > failure will be EACCES, as your report shows.  That's because the
> > default way of opening files which we use in Emacs opens them in
> > "unshared" mode, so they cannot be shared between different processes.
> 
> But in this case, Emacs 2 should have exited already before we try to
> read the file...  I think.  I'll try to debug this more thoroughly;
> perhaps I was misinterpreting what I was seeing.
> 
> > Which probably means multisession.el should have some mechanism to
> > deal with this gracefully, like retrying the open/read several times
> > before giving up.  After all, the window for such races should be
> > quite short, given that we write very small files.
> 
> Yup.  It does this in the sqlite case, so I'll replicate the logic in
> the files case, too.  But it seemed very mysterious to me that I was
> getting a permission denied after Emacs 2 had exited.

If you are sure the second session already exited, then my hand-waving
doesn't hold water.