unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
* Re: autosave
       [not found] <200206080100.g5810Sbw007341@mail.nj.nec.com>
@ 2002-06-09 15:19 ` Richard Stallman
  2002-06-09 16:53   ` autosave Peter Blicher
  2002-06-09 17:05   ` autosave Peter Blicher
  0 siblings, 2 replies; 9+ messages in thread
From: Richard Stallman @ 2002-06-09 15:19 UTC (permalink / raw)
  Cc: emacs-devel

    When unable to create (or write to?) some autosave dir .emacs.d or below,
    retries forever.  This occurred when I was running with a home dir that I
    couldn't write to (because I was root on a different machine).  Never had
    this problem in versions 20.4 and before.

Would you please supply a precise test case so we can reproduce this?
Please read the Bugs section in the Emacs manual, which provides
guidelines on how to write a bug report to give us the
necessary information so we can fix the bug.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-09 15:19 ` autosave Richard Stallman
@ 2002-06-09 16:53   ` Peter Blicher
  2002-06-10 23:43     ` autosave Richard Stallman
  2002-06-09 17:05   ` autosave Peter Blicher
  1 sibling, 1 reply; 9+ messages in thread
From: Peter Blicher @ 2002-06-09 16:53 UTC (permalink / raw)
  Cc: emacs-devel

   Date: Sun, 9 Jun 2002 09:19:03 -0600 (MDT)
   X-Authentication-Warning: aztec.santafe.edu: rms set sender to rms@aztec using -f
   From: Richard Stallman <rms@gnu.org>
   cc: emacs-devel@gnu.org
   Reply-to: rms@gnu.org
   Content-Length: 516

       When unable to create (or write to?) some autosave dir .emacs.d or below,
       retries forever.  This occurred when I was running with a home dir that I
       couldn't write to (because I was root on a different machine).  Never had
       this problem in versions 20.4 and before.

   Would you please supply a precise test case so we can reproduce this?
   Please read the Bugs section in the Emacs manual, which provides
   guidelines on how to write a bug report to give us the
   necessary information so we can fix the bug.

I've read the bugs sections many times over many years.  This was the best I
could do under the circumstances.  If I get time, I will try to get you
better info.  

However, if someone sent me such a bug report, what I would do is take a look
at the code that does auto-save, and see what it does when it tries to create
or access the directories mentioned.  If that code has changed since the
previous major release, I would take a closer look.  I would consider what
would happen if it did not have adequate permissions for places it wanted to
write to or create or check the existence of directories in.

--peter

-- 
A. Peter Blicher        NEC Research Institute
609-951-2716            4 Independence Way
609-951-2488 fax        Princeton, NJ  08540-6634  USA
blicher@research.nj.nec.com

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-09 15:19 ` autosave Richard Stallman
  2002-06-09 16:53   ` autosave Peter Blicher
@ 2002-06-09 17:05   ` Peter Blicher
  2002-06-10 23:44     ` autosave Richard Stallman
  1 sibling, 1 reply; 9+ messages in thread
From: Peter Blicher @ 2002-06-09 17:05 UTC (permalink / raw)
  Cc: emacs-devel

   Date: Sun, 9 Jun 2002 09:19:03 -0600 (MDT)
   X-Authentication-Warning: aztec.santafe.edu: rms set sender to rms@aztec using -f
   From: Richard Stallman <rms@gnu.org>
   cc: emacs-devel@gnu.org
   Reply-to: rms@gnu.org
   Content-Length: 516

       When unable to create (or write to?) some autosave dir .emacs.d or below,
       retries forever.  This occurred when I was running with a home dir that I
       couldn't write to (because I was root on a different machine).  Never had
       this problem in versions 20.4 and before.

   Would you please supply a precise test case so we can reproduce this?
   Please read the Bugs section in the Emacs manual, which provides
   guidelines on how to write a bug report to give us the
   necessary information so we can fix the bug.

Oh, I guess I should mention a couple of other things:

1.  I changed the permissions on the .emacs.d dir to be rwx everyone, and
    that fixed the problem.  (Thus lending credence to my theory that it
    can't figure out that it can't write and/or read that dir.

2.  I think I recall that in the past, when emacs could not for some reason
    write an autosave file, it would simply issue a message to that effect
    and go on, rather than get into an infinite loop of trying to access the
    autosave directory.  If I'm not mistaken, the autosave files used to be
    written to the home dir; now apparently there is a new separate dir
    created in the home dir for that purpose.  While this leaves less of a
    mess in the home dir, it makes me think it is highly likely that someone
    cleaned up some code a little too much.

3.  Since I have like 20 years of historic emacs lisp code I am loading some
    of, which generally has to be trimmed at each new major release, it is
    true that this problem could somehow be my fault.  Alas, I don't have
    time right now to check that.

4.  Though I don't have time to do this myself right now, what I would
    suggest to try to reproduce this bug is this: Run with a home dir that
    contains an .emacs.d directory that you do not have any of rwx on.  Put
    emacs into a state where it wants to autosave something.  See what
    happens.  If you don't have a problem, email me and say you can't do
    anything until I give you better info.  Else, fix the bug.

--peter

-- 
A. Peter Blicher        NEC Research Institute
609-951-2716            4 Independence Way
609-951-2488 fax        Princeton, NJ  08540-6634  USA
blicher@research.nj.nec.com

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-09 16:53   ` autosave Peter Blicher
@ 2002-06-10 23:43     ` Richard Stallman
  0 siblings, 0 replies; 9+ messages in thread
From: Richard Stallman @ 2002-06-10 23:43 UTC (permalink / raw)
  Cc: emacs-devel

    However, if someone sent me such a bug report, what I would do is take a look
    at the code that does auto-save, and see what it does when it tries to create
    or access the directories mentioned.

There is a lot of code for auto-saving.  Before I wrote to you, I
looked through it but I didn't see anything that tries to create auto
save files in an emacs.d directory there.  Unless I can reproduce it,
I don't see how to do any more.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-09 17:05   ` autosave Peter Blicher
@ 2002-06-10 23:44     ` Richard Stallman
  2002-06-11  4:39       ` autosave Eli Zaretskii
  0 siblings, 1 reply; 9+ messages in thread
From: Richard Stallman @ 2002-06-10 23:44 UTC (permalink / raw)
  Cc: emacs-devel

    Oh, I guess I should mention a couple of other things:

Actually the one thing I would like is a complete self-contained
recipe for making the bug happen.  It is not normal for Emacs to write
auto-save files in emacs.d.

      If I'm not mistaken, the autosave files used to be
	written to the home dir;

Not normally.

     now apparently there is a new separate dir
	created in the home dir for that purpose.

Not normally.  So I think you must have done something unusual.  That
is why I need a precise recipe for how to make this happen.

    4.  Though I don't have time to do this myself right now, what I would
	suggest to try to reproduce this bug is this: Run with a home dir that
	contains an .emacs.d directory that you do not have any of rwx on.  Put
	emacs into a state where it wants to autosave something.  See what
	happens.

What happens is it works fine and puts these files in the usual place,
which is not in emacs.d.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-10 23:44     ` autosave Richard Stallman
@ 2002-06-11  4:39       ` Eli Zaretskii
  2002-06-11 18:10         ` autosave Peter Blicher
  2002-06-12  2:31         ` autosave Richard Stallman
  0 siblings, 2 replies; 9+ messages in thread
From: Eli Zaretskii @ 2002-06-11  4:39 UTC (permalink / raw)
  Cc: blicher, emacs-devel


On Mon, 10 Jun 2002, Richard Stallman wrote:

>     Oh, I guess I should mention a couple of other things:
> 
> Actually the one thing I would like is a complete self-contained
> recipe for making the bug happen.  It is not normal for Emacs to write
> auto-save files in emacs.d.

The ~/.emacs.d/auto-save-list/ directory is used for the `.saves-*' files 
where Emacs saves session information (used by "M-x recover-session").  
(See startup.el:auto-save-list-file-prefix.)

These are not the normal auto-save files, but they do resemble them 
because Emacs saves session info periodically, like it does with 
auto-saving.  Perhaps the problem was with those files, not with autosave 
files.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-11  4:39       ` autosave Eli Zaretskii
@ 2002-06-11 18:10         ` Peter Blicher
  2002-06-12  4:46           ` autosave Eli Zaretskii
  2002-06-12  2:31         ` autosave Richard Stallman
  1 sibling, 1 reply; 9+ messages in thread
From: Peter Blicher @ 2002-06-11 18:10 UTC (permalink / raw)
  Cc: rms, emacs-devel

   Date: Tue, 11 Jun 2002 07:39:27 +0300 (IDT)
   From: Eli Zaretskii <eliz@is.elta.co.il>
   X-Sender: eliz@is
   cc: blicher@research.nj.nec.com, emacs-devel@gnu.org
   Content-Type: TEXT/PLAIN; charset=US-ASCII
   Content-Length: 693


   On Mon, 10 Jun 2002, Richard Stallman wrote:

   >     Oh, I guess I should mention a couple of other things:
   > 
   > Actually the one thing I would like is a complete self-contained
   > recipe for making the bug happen.  It is not normal for Emacs to write
   > auto-save files in emacs.d.

   The ~/.emacs.d/auto-save-list/ directory is used for the `.saves-*' files 
   where Emacs saves session information (used by "M-x recover-session").  
   (See startup.el:auto-save-list-file-prefix.)

   These are not the normal auto-save files, but they do resemble them 
   because Emacs saves session info periodically, like it does with 
   auto-saving.  Perhaps the problem was with those files, not with autosave 
   files.

That's highly likely.  However, given that rms has said he already looked at
the code, I think he is right that I will have to isolate the problem better
before you guys can do more.  Unless he did not look at the session save
code.

Unfortunately, it is hard to look at something like the *Messages* buffer
when the program is in an infinite loop of failed file operations.  Since
it's been years since I did emacs debugging, I don't remember offhand all the
stuff I have to do to debug, trace, etc., which means I have to look all this
stuff up before I can give you a distilled version of the problem.

--peer

-- 
A. Peter Blicher        NEC Research Institute
609-951-2716            4 Independence Way
609-951-2488 fax        Princeton, NJ  08540-6634  USA
blicher@research.nj.nec.com

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-11  4:39       ` autosave Eli Zaretskii
  2002-06-11 18:10         ` autosave Peter Blicher
@ 2002-06-12  2:31         ` Richard Stallman
  1 sibling, 0 replies; 9+ messages in thread
From: Richard Stallman @ 2002-06-12  2:31 UTC (permalink / raw)
  Cc: blicher, emacs-devel

    The ~/.emacs.d/auto-save-list/ directory is used for the `.saves-*' files 
    where Emacs saves session information (used by "M-x recover-session").  
    (See startup.el:auto-save-list-file-prefix.)

    These are not the normal auto-save files, but they do resemble them 
    because Emacs saves session info periodically, like it does with 
    auto-saving.  Perhaps the problem was with those files, not with autosave 
    files.

I looked at that code.  If Emacs can't open that file,
it simply does not write that file.  It does not retry.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: autosave
  2002-06-11 18:10         ` autosave Peter Blicher
@ 2002-06-12  4:46           ` Eli Zaretskii
  0 siblings, 0 replies; 9+ messages in thread
From: Eli Zaretskii @ 2002-06-12  4:46 UTC (permalink / raw)
  Cc: emacs-devel


On Tue, 11 Jun 2002, Peter Blicher wrote:

> Since
> it's been years since I did emacs debugging, I don't remember offhand all the
> stuff I have to do to debug, trace, etc., which means I have to look all this
> stuff up before I can give you a distilled version of the problem.

The file etc/DEBUG in the distribution should get you off the ground.

^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2002-06-12  4:46 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <200206080100.g5810Sbw007341@mail.nj.nec.com>
2002-06-09 15:19 ` autosave Richard Stallman
2002-06-09 16:53   ` autosave Peter Blicher
2002-06-10 23:43     ` autosave Richard Stallman
2002-06-09 17:05   ` autosave Peter Blicher
2002-06-10 23:44     ` autosave Richard Stallman
2002-06-11  4:39       ` autosave Eli Zaretskii
2002-06-11 18:10         ` autosave Peter Blicher
2002-06-12  4:46           ` autosave Eli Zaretskii
2002-06-12  2:31         ` autosave Richard Stallman

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).