WARNING: The attached message make Gnus eat all resources

WARNING: The attached message make Gnus eat all resources

[Nevin Kapur]

[-- Attachment #1: Type: text/plain, Size: 333 bytes --]

The attached message will make Gnus eat all resources for a *long*
time -- I C-g'd after about a minute.  Debug on quit suggests
something to do with flow-fill and filladapt in turn.

On an unrelated note, when I tried to forward the message with C-c
C-f, it stripped out the Content-Type header which made the bug go
away.

-Nevin


[-- Attachment #2: Type: message/rfc822, Size: 2750 bytes --]

From: John Pettitt <jpp@cloudview.com>
Subject: Re: [Admin/Spamassasin] Re: PHP Security Advisory: Vulnerability in PHP versions 4.2.0 and 4.2.1
Date: Mon, 22 Jul 2002 20:35:30 -0700
Message-ID: <5.1.0.14.2.20020722203245.00b5a868@gatekeeper.cloudview.com>




>Will people on this list please refrain from putting other people's
>mail on Razor? It's *not* helpful when important announcements end
>up in the bit bucket, to be found when the bucket's owner gets
>around to empty it.

Which translates as - if you want to stop people from seeing security
advisories submit them to razor.  Perhaps would be better to configure
spam filters to ignore bugtraq so as to avoid this kind of attack.



John Pettitt                                     Email: jpp@cloudview.com

"Do what you feel in your heart to be right for you'll be criticized anyway."
                                                                                       - 
Eleanor Roosevelt



[-- Attachment #3: Type: text/plain, Size: 10 bytes --]




-Nevin

Re: WARNING: The attached message make Gnus eat all resources

[Reiner Steib]

On Tue, Jul 23 2002, Nevin Kapur wrote:

> The attached message will make Gnus eat all resources for a *long*
> time -- I C-g'd after about a minute.  Debug on quit suggests
> something to do with flow-fill and filladapt in turn.

I can display this message (directly in gmane.comp.security.bugtraq)
without any problems. I use GNU Emacs 21.1.1, Oort Gnus v0.07. (I have
no special setting related to filling.)

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo--- PGP key available via WWW   http://rsteib.home.pages.de/

Re: WARNING: The attached message make Gnus eat all resources

[Josh Huber]

Reiner Steib <4uce.02.r.steib@gmx.net> writes:

> I can display this message (directly in gmane.comp.security.bugtraq)
> without any problems. I use GNU Emacs 21.1.1, Oort Gnus v0.07. (I
> have no special setting related to filling.)

It eats all my CPU. Gnus v0.07, XEmacs 21.4.8

-- 
Josh Huber

Re: WARNING: The attached message make Gnus eat all resources

[Harry Putnam]

Josh Huber <huber+dated+1027879739.c41f0b@alum.wpi.edu> writes:

> Reiner Steib <4uce.02.r.steib@gmx.net> writes:
>
>> I can display this message (directly in gmane.comp.security.bugtraq)
>> without any problems. I use GNU Emacs 21.1.1, Oort Gnus v0.07. (I
>> have no special setting related to filling.)
>
> It eats all my CPU. Gnus v0.07, XEmacs 21.4.8

Nails mine too ..Gnus v0.07 emacs-21.3.50

Re: WARNING: The attached message make Gnus eat all resources

[Niklas Morberg]

Nevin Kapur <nevin@jhu.edu> writes:

> The attached message will make Gnus eat all resources for
> a *long* time -- I C-g'd after about a minute. Debug on
> quit suggests something to do with flow-fill and filladapt
> in turn.

Seems like a correct assumption. It eats all my CPU on
latest CVS Gnus, GNU Emacs 21.2.1 (i386-mingw-nt5.0.2195).
Turning filladapt-mode off makes it possible to view the
article without any problems.

I'm using filladapt 2.12. Is there a later version? Perhaps
one that understands the ChangeLog format as well?

Niklas

Re: WARNING: The attached message make Gnus eat all resources

[Simon Josefsson]

Nevin Kapur <nevin@jhu.edu> writes:

> The attached message will make Gnus eat all resources for a *long*
> time -- I C-g'd after about a minute.  Debug on quit suggests
> something to do with flow-fill and filladapt in turn.
>
> On an unrelated note, when I tried to forward the message with C-c
> C-f, it stripped out the Content-Type header which made the bug go
> away.

Does this fix it?

Fill-adapt is IMHO evil as it redefines existing elisp functions.

--- flow-fill.el.~6.11.~	2002-06-09 21:47:48.000000000 +0200
+++ flow-fill.el	2002-07-08 15:30:31.000000000 +0200
@@ -135,7 +135,8 @@
 	  (unless sig
 	    (condition-case nil
 		(let ((fill-prefix (when quote (concat quote " ")))
-		      (fill-column (eval fill-flowed-display-column)))
+		      (fill-column (eval fill-flowed-display-column))
+		      filladapt-mode)
 		  (fill-region (fill-flowed-point-at-bol)
 			       (min (1+ (fill-flowed-point-at-eol)) (point-max))
 			       'left 'nosqueeze))

Re: WARNING: The attached message make Gnus eat all resources

[Nevin Kapur]

Simon Josefsson <jas@extundo.com> writes:

> Nevin Kapur <nevin@jhu.edu> writes:
>
>> The attached message will make Gnus eat all resources for a *long*
>> time -- I C-g'd after about a minute.  Debug on quit suggests
>> something to do with flow-fill and filladapt in turn.

[...]

> Does this fix it?

Nope.

> Fill-adapt is IMHO evil as it redefines existing elisp functions.

I turned off filladapt and I still see the bug.  Here's the backtrace
when I quite.  I can't see anything useful in it, though.

Signaling: (quit)
  move-to-left-margin()
  forward-paragraph(1)
  fill-region(3611 3701 left nosqueeze)
  (let ((fill-prefix ...) (fill-column ...) filladapt-mode) (fill-region (fill-flowed-point-at-bol) (min ... ...) (quote left) (quote nosqueeze)))
  (condition-case nil (let (... ... filladapt-mode) (fill-region ... ... ... ...)) (error nil))
  (if sig nil (condition-case nil (let ... ...) (error nil)))
  (unless sig (condition-case nil (let ... ...) (error nil)))
  (let ((quote ...) sig) (if (string= quote "") (setq quote nil)) (when (and quote ...) (save-excursion ... ...)) (while (and ... ...) (save-excursion ...) (backward-delete-char -1) (end-of-line)) (unless sig (condition-case nil ... ...)))
  (if (save-excursion (beginning-of-line) (looking-at "^\\(>*\\)\\( ?\\)")) (let (... sig) (if ... ...) (when ... ...) (while ... ... ... ...) (unless sig ...)))
  (when (save-excursion (beginning-of-line) (looking-at "^\\(>*\\)\\( ?\\)")) (let (... sig) (if ... ...) (when ... ...) (while ... ... ... ...) (unless sig ...)))
  (while (re-search-forward " $" nil t) (when (save-excursion ... ...) (let ... ... ... ... ...)))
  (save-excursion (set-buffer (or ... buffer)) (goto-char (point-min)) (while (re-search-forward " $" nil t) (when ... ...)))
  fill-flowed()
  mm-inline-text((#<buffer " *mm*<6>"> ("text/plain" (charset . "us-ascii") (format . "flowed")) nil nil nil nil nil nil))
  mm-display-inline((#<buffer " *mm*<6>"> ("text/plain" (charset . "us-ascii") (format . "flowed")) nil nil nil nil nil nil))
  mm-display-part((#<buffer " *mm*<6>"> ("text/plain" (charset . "us-ascii") (format . "flowed")) nil nil nil nil nil nil) t)
  byte-code("..." [not-attachment type ignored display beg handle string-match throw nil mm-inline-override-p 4 "inline" mm-attachment-override-p mm-automatic-display-p mm-inlinable-p mm-inlined-p mm-automatic-external-display-p t split-string mm-handle-media-type "/" "text" gnus-unbuttonized-mime-type-p gnus-insert-mime-button gnus-article-insert-newline -1 (set-buffer gnus-summary-buffer) ((error)) mm-display-part mm-insert-inline mm-get-part gnus-treat-article bufferp 1 text gnus-article-mime-handle-alist id move gnus-article-mime-handles gnus-newsgroup-charset gnus-newsgroup-ignored-charsets mail-parse-ignored-charsets mail-parse-charset] 7)
  gnus-mime-display-single((#<buffer " *mm*<6>"> ("text/plain" (charset . "us-ascii") (format . "flowed")) nil nil nil nil nil nil))
  gnus-mime-display-part((#<buffer " *mm*<6>"> ("text/plain" (charset . "us-ascii") (format . "flowed")) nil nil nil nil nil nil))
  gnus-display-mime()
  gnus-article-prepare-display()
  mm-inline-message((#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil))
  mm-display-inline((#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil))
  mm-display-part((#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil) t)
  byte-code("..." [not-attachment type ignored display beg handle string-match throw nil mm-inline-override-p 4 "inline" mm-attachment-override-p mm-automatic-display-p mm-inlinable-p mm-inlined-p mm-automatic-external-display-p t split-string mm-handle-media-type "/" "text" gnus-unbuttonized-mime-type-p gnus-insert-mime-button gnus-article-insert-newline -1 (set-buffer gnus-summary-buffer) ((error)) mm-display-part mm-insert-inline mm-get-part gnus-treat-article bufferp 1 text gnus-article-mime-handle-alist id move gnus-article-mime-handles gnus-newsgroup-charset gnus-newsgroup-ignored-charsets mail-parse-ignored-charsets mail-parse-charset] 7)
  gnus-mime-display-single((#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil))
  gnus-mime-display-part((#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil))
  mapcar(gnus-mime-display-part ((#<buffer " *mm*<2>"> ("text/plain") nil (lambda nil ...) nil nil nil nil) (#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil) (#<buffer " *mm*<4>"> ("text/plain") nil nil nil nil nil nil)))
  gnus-mime-display-mixed(((#<buffer " *mm*<2>"> ("text/plain") nil (lambda nil ...) nil nil nil nil) (#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil) (#<buffer " *mm*<4>"> ("text/plain") nil nil nil nil nil nil)))
  gnus-mime-display-part(("multipart/mixed" (#<buffer " *mm*<2>"> ("text/plain") nil (lambda nil ...) nil nil nil nil) (#<buffer " *mm*<3>"> ("message/rfc822") nil nil ("inline") nil nil nil) (#<buffer " *mm*<4>"> ("text/plain") nil nil nil nil nil nil)))
  gnus-display-mime()
  gnus-article-prepare-display()
  gnus-article-prepare(1823 nil)
  gnus-summary-display-article(1823 nil)
  gnus-summary-select-article(nil force)
  gnus-summary-show-article(nil)
  call-interactively(gnus-summary-show-article)

-Nevin

Re: WARNING: The attached message make Gnus eat all resources

[Simon Josefsson]

Nevin Kapur <nevin@jhu.edu> writes:

>> Fill-adapt is IMHO evil as it redefines existing elisp functions.
>
> I turned off filladapt and I still see the bug.  Here's the backtrace
> when I quite.  I can't see anything useful in it, though.
>
> Signaling: (quit)
>   move-to-left-margin()
>   forward-paragraph(1)
>   fill-region(3611 3701 left nosqueeze)
>   (let ((fill-prefix ...) (fill-column ...) filladapt-mode) (fill-region (fill-flowed-point-at-bol) (min ... ...) (quote left) (quote nosqueeze)))
>   (condition-case nil (let (... ... filladapt-mode) (fill-region ... ... ... ...)) (error nil))

Is this XEmacs specific?  The message works fine under Emacs here.
I'll try XEmacs soon.

Re: WARNING: The attached message make Gnus eat all resources

[Nevin Kapur]

Simon Josefsson <jas@extundo.com> writes:

> Nevin Kapur <nevin@jhu.edu> writes:
>
>>> Fill-adapt is IMHO evil as it redefines existing elisp functions.
>>
>> I turned off filladapt and I still see the bug.  Here's the backtrace
>> when I quite.  I can't see anything useful in it, though.
>>
>> Signaling: (quit)
>>   move-to-left-margin()
>>   forward-paragraph(1)
>>   fill-region(3611 3701 left nosqueeze)
>>   (let ((fill-prefix ...) (fill-column ...) filladapt-mode) (fill-region (fill-flowed-point-at-bol) (min ... ...) (quote left) (quote nosqueeze)))
>>   (condition-case nil (let (... ... filladapt-mode) (fill-region ... ... ... ...)) (error nil))
>
> Is this XEmacs specific?  The message works fine under Emacs here.
> I'll try XEmacs soon.

Somebody else reported that they had a problem with CVS Emacs too but
they claimed that turning off filladapt made the problem go away.

-Nevin

Re: WARNING: The attached message make Gnus eat all resources

[Lars Magne Ingebrigtsen]

Simon Josefsson <jas@extundo.com> writes:

> Is this XEmacs specific?  The message works fine under Emacs here.
> I'll try XEmacs soon.

The problem was with the error handling code -- `fill-prefix' was too
long, which made the `condition-case' take the error, but not advance
point.  I've just added a `forward-line' to the error branch.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: WARNING: The attached message make Gnus eat all resources

[Lars Magne Ingebrigtsen]

Nevin Kapur <nevin@jhu.edu> writes:

> The attached message will make Gnus eat all resources for a *long*
> time -- I C-g'd after about a minute.  Debug on quit suggests
> something to do with flow-fill and filladapt in turn.

Yup.  Fix in Oort Gnus v0.08 (i. e., CVS).

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen