bug#24307: GNU Emacs 25.1.1 repeated system crash

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

GNU Emacs 25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.14.5) of 2016-08-22
Linux [...] 3.16.0-4-686-pae #1 SMP Debian 3.16.7-ckt25-2+deb8u3 
(2016-07-02) i686 GNU/Linux

Last days recent pretest caused two times a system freeze, restart required.
Unfortunately  couldn't reproduce with Emacs -Q
As this didn't happen last year --but sometimes the years before-- maybe 
it's a valuable info though.
A crash happened after M-x edit-abbrevs RET, editing

"gna"           0    "(< 0 (abs"
"wgna"           0    "(when (< 0 (abs"

bug#24307: GNU Emacs 25.1.1 repeated system crash

[John Wiegley]

>>>>> "AR" == Andreas Röhler <andreas.roehler@easy-emacs.de> writes:

AR> Last days recent pretest caused two times a system freeze, restart
AR> required.

A full system freeze? That's pretty impressive. Was it maybe a runaway
allocation causing the system to swap?

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 25.08.2016 09:04, John Wiegley wrote:
>>>>>> "AR" == Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
> AR> Last days recent pretest caused two times a system freeze, restart
> AR> required.
>
> A full system freeze? That's pretty impressive. Was it maybe a runaway
> allocation causing the system to swap?
>

The edit-abbrevs has a design-issue, dealing with all when changing one. 
Maybe that turned exponential?

bug#24307: GNU Emacs 25.1.1 repeated system crash

[John Wiegley]

>>>>> Andreas Röhler <andreas.roehler@easy-emacs.de> writes:

> The edit-abbrevs has a design-issue, dealing with all when changing one.
> Maybe that turned exponential?

But even if so, that would only lock up a single core -- unless it was busily
trying to create an abbrevs file as big as all available disk space, while at
the same time filling memory...

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 25.08.2016 09:21, John Wiegley wrote:
>>>>>> Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
>> The edit-abbrevs has a design-issue, dealing with all when changing one.
>> Maybe that turned exponential?
> But even if so, that would only lock up a single core -- unless it was busily
> trying to create an abbrevs file as big as all available disk space, while at
> the same time filling memory...
>

The other form causing a restart was navigation-related inside an 
incomplete form of

(when (setq VAR (FUNKTION ARG))

Probably not really helpful at the moment, just wanted to get it known,

Cheers,

Andreas

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Eli Zaretskii]

> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
> Date: Thu, 25 Aug 2016 09:01:56 +0200
> 
> GNU Emacs 25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.14.5) of 2016-08-22
> Linux [...] 3.16.0-4-686-pae #1 SMP Debian 3.16.7-ckt25-2+deb8u3 
> (2016-07-02) i686 GNU/Linux
> 
> Last days recent pretest caused two times a system freeze, restart required.
> Unfortunately  couldn't reproduce with Emacs -Q
> As this didn't happen last year --but sometimes the years before-- maybe 
> it's a valuable info though.
> A crash happened after M-x edit-abbrevs RET, editing
> 
> "gna"           0    "(< 0 (abs"
> "wgna"           0    "(when (< 0 (abs"

Can you provide more details about these system freezes?  Some
questions:

 . What exactly freezes?  Can you start a new application, or move the
   mouse pointer, or ping the system from the net?

 . What do you see in /var/log/messages around the time of these
   freezes?

 . When it freezes, do you see the disk I/O indicator lit, indicating
   some disk I/O, or is the indicator off?

 . Any other relevant details.

Thanks.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 25.08.2016 16:38, Eli Zaretskii wrote:
>> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
>> Date: Thu, 25 Aug 2016 09:01:56 +0200
>>
>> GNU Emacs 25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.14.5) of 2016-08-22
>> Linux [...] 3.16.0-4-686-pae #1 SMP Debian 3.16.7-ckt25-2+deb8u3
>> (2016-07-02) i686 GNU/Linux
>>
>> Last days recent pretest caused two times a system freeze, restart required.
>> Unfortunately  couldn't reproduce with Emacs -Q
>> As this didn't happen last year --but sometimes the years before-- maybe
>> it's a valuable info though.
>> A crash happened after M-x edit-abbrevs RET, editing
>>
>> "gna"           0    "(< 0 (abs"
>> "wgna"           0    "(when (< 0 (abs"
> Can you provide more details about these system freezes?  Some
> questions:
>
>   . What exactly freezes?  Can you start a new application, or move the
>     mouse pointer, or ping the system from the net?
>
>   . What do you see in /var/log/messages around the time of these
>     freezes?
>
>   . When it freezes, do you see the disk I/O indicator lit, indicating
>     some disk I/O, or is the indicator off?
>
>   . Any other relevant details.
>
> Thanks.

Hi Eli,

system was completely frozen, no message, no mouse, no reaction at all.

At the time thunderbird was open too. Maybe it was caused by an attack 
from there.

Thanks,
Andreas

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Eli Zaretskii]

> Cc: 24307@debbugs.gnu.org
> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
> Date: Thu, 25 Aug 2016 17:27:11 +0200
> 
> system was completely frozen, no message, no mouse, no reaction at all.

Then what you find in /var/log/messages around that time may be the
only clue to what happened.  Please see if you find anything
interesting there.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 25.08.2016 18:20, Eli Zaretskii wrote:
>> Cc: 24307@debbugs.gnu.org
>> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
>> Date: Thu, 25 Aug 2016 17:27:11 +0200
>>
>> system was completely frozen, no message, no mouse, no reaction at all.
> Then what you find in /var/log/messages around that time may be the
> only clue to what happened.  Please see if you find anything
> interesting there.

Hmm, looks dbus related -  the part of the messages have been sent 
off-list.
Could run Emacs in some debug mode.
Any suggestions? valgrind exists here.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Eli Zaretskii]

> Cc: 24307@debbugs.gnu.org
> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
> Date: Thu, 25 Aug 2016 19:24:52 +0200
> 
> Hmm, looks dbus related -  the part of the messages have been sent 
> off-list.

No, I don't think dbus is related, its message just says that the
system is being shut down.

> Could run Emacs in some debug mode.
> Any suggestions? valgrind exists here.

If the system freezes, how can a debugger help?  You will be unable to
get any response from the debugger.  Or does something remain unfrozen
after all?

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Mike Kupfer]

Hi Andreas, do you have a second system that you can use to (try to) ssh
into the hung system, or at least ping it?

Earlier in the year I had a reproducible system hang that was apparently
related to a video driver lockup.  The problem went away if I switched
from MATE to Xfce, or if I disabled thumbnails in the MATE alt-tab
window switcher.  A couple days later there was a kernel update that
included a fix for the video driver, and after that I could no longer
reproduce the hang.

What sort of desktop environment are you using?

regards,
mike

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 25.08.2016 19:34, Eli Zaretskii wrote:
>> Cc: 24307@debbugs.gnu.org
>> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
>> Date: Thu, 25 Aug 2016 19:24:52 +0200
>>
>> Hmm, looks dbus related -  the part of the messages have been sent
>> off-list.
> No, I don't think dbus is related, its message just says that the
> system is being shut down.
>
>> Could run Emacs in some debug mode.
>> Any suggestions? valgrind exists here.
> If the system freezes, how can a debugger help?  You will be unable to
> get any response from the debugger.

Thought at some logging, dumping.

>   Or does something remain unfrozen
> after all?

Not in this both recent cases.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 25.08.2016 19:53, Mike Kupfer wrote:
> Hi Andreas, do you have a second system that you can use to (try to) ssh
> into the hung system, or at least ping it?

Yes, have a second one, Xubuntu. Could try that.

> Earlier in the year I had a reproducible system hang that was apparently
> related to a video driver lockup.  The problem went away if I switched
> from MATE to Xfce, or if I disabled thumbnails in the MATE alt-tab
> window switcher.  A couple days later there was a kernel update that
> included a fix for the video driver, and after that I could no longer
> reproduce the hang.
>
> What sort of desktop environment are you using?
>
> regards,
> mike

Debian, Xfce.

Thanks,
Andreas

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Nicolas Richard]

Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
> system was completely frozen, no message, no mouse, no reaction at all.
>
> At the time thunderbird was open too. Maybe it was caused by an attack
> from there.

Did you try https://en.wikipedia.org/wiki/Magic_SysRq_key ? (sometimes
they are disabled though)

Nicolas.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 26.08.2016 13:17, Nicolas Richard wrote:
> Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
>> system was completely frozen, no message, no mouse, no reaction at all.
>>
>> At the time thunderbird was open too. Maybe it was caused by an attack
>> from there.
> Did you try https://en.wikipedia.org/wiki/Magic_SysRq_key ? (sometimes
> they are disabled though)
>
> Nicolas.

/sbin/sysctl -a gives me a value  for kernel.sysrq
However a key like

Alt+SysRQ+m

which should print memory information to the console.

according to

http://linuxhowtos.org/Tips%20and%20Tricks/sysrq.htm

seems to have no effect.

SysRQ understood as the print-key.

Thanks,

Andreas

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Nicolas Richard]

Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
> Alt+SysRQ+m
>
> seems to have no effect.
>
> SysRQ understood as the print-key.

WFM, but perhaps your sysrq key is elsewhere ?

Also check if it appears in your dmesg output : sometimes the logging
level of your console simply won't show the sysrq messages.

If that doesn't work I'm afraid I don't have any further suggestion.

Nicolas.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

Got it, thanks!

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

[-- Attachment #1: Type: text/plain, Size: 3015 bytes --]

This morning the keyboard messed up X-wide - backspace in X didn't work 
any more.

Wrote the following into some buffer in sh-mode:

${WERKSTATT=1}

See attached png what was displayed, when trying to M-x report-emacs-bug 
from emacs -Q

Send the results of M-x report-emacs-bug after restart, which might 
provide some info still.


In GNU Emacs 25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.14.5)
  of 2016-08-22 built on sprache
Windowing system distributor 'The X.Org Foundation', version 11.0.11604000
System Description:    Debian GNU/Linux 8.5 (jessie)

Configured features:
XPM JPEG TIFF GIF PNG RSVG SOUND GPM DBUS GSETTINGS NOTIFY ACL LIBXML2
FREETYPE M17N_FLT LIBOTF XFT ZLIB TOOLKIT_SCROLL_BARS GTK3 X11

Important settings:
   value of $LANG: de_DE.utf8
   value of $XMODIFIERS: @im=ibus
   locale-coding-system: utf-8-unix

Major mode: Lisp Interaction

Minor modes in effect:
   tooltip-mode: t
   global-eldoc-mode: t
   electric-indent-mode: t
   mouse-wheel-mode: t
   tool-bar-mode: t
   menu-bar-mode: t
   file-name-shadow-mode: t
   global-font-lock-mode: t
   font-lock-mode: t
   blink-cursor-mode: t
   auto-composition-mode: t
   auto-encryption-mode: t
   auto-compression-mode: t
   line-number-mode: t
   transient-mark-mode: t

Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.
Loading 
/home/speck/.emacs.d/session.2f82803a4-a0e9-4eb9-8a7b-41136a5e202b...done
Restored session data

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message dired format-spec rfc822 mml
mml-sec password-cache epg epg-config gnus-util mm-decode mm-bodies
mm-encode mail-parse rfc2231 mailabbrev gmm-utils mailheader sendmail
rfc2047 rfc2045 ietf-drums mm-util help-fns help-mode easymenu
cl-loaddefs pcase cl-lib mail-prsvr mail-utils time-date mule-util
tooltip eldoc electric uniquify ediff-hook vc-hooks lisp-float-type
mwheel x-win term/common-win x-dnd tool-bar dnd fontset image regexp-opt
fringe tabulated-list newcomment elisp-mode lisp-mode prog-mode register
page menu-bar rfn-eshadow timer select scroll-bar mouse jit-lock
font-lock syntax facemenu font-core frame cl-generic cham georgian
utf-8-lang misc-lang vietnamese tibetan thai tai-viet lao korean
japanese eucjp-ms cp51932 hebrew greek romanian slovak czech european
ethiopic indian cyrillic chinese charscript case-table epa-hook
jka-cmpr-hook help simple abbrev minibuffer cl-preloaded nadvice
loaddefs button faces cus-face macroexp files text-properties overlay
sha1 md5 base64 format env code-pages mule custom widget
hashtable-print-readable backquote dbusbind inotify dynamic-setting
system-font-setting font-render-setting move-toolbar gtk x-toolkit x
multi-tty make-network-process emacs)

Memory information:
((conses 8 159575 7353)
  (symbols 24 38009 0)
  (miscs 20 45 138)
  (strings 16 50161 4964)
  (string-bytes 1 716907)
  (vectors 8 12699)
  (vector-slots 4 462985 5518)
  (floats 8 166 63)
  (intervals 28 187 14)
  (buffers 520 18)
  (heap 1024 19124 913))


[-- Attachment #2: dollar.png --]
[-- Type: image/png, Size: 70924 bytes --]

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Eli Zaretskii]

> Cc: 24307@debbugs.gnu.org, Eli Zaretskii <eliz@gnu.org>,
>  John Wiegley <jwiegley@gmail.com>
> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
> Date: Sun, 28 Aug 2016 09:36:55 +0200
> 
> This morning the keyboard messed up X-wide - backspace in X didn't work 
> any more.
> 
> Wrote the following into some buffer in sh-mode:
> 
> ${WERKSTATT=1}
> 
> See attached png what was displayed, when trying to M-x report-emacs-bug 
> from emacs -Q

Looks like your Super key is stuck pressed?  Otherwise, why does the
image show a message about M-s-x being unbound?

Other than that, I see nothing of interest in the image you posted.
Did I miss something?

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 28.08.2016 16:28, Eli Zaretskii wrote:
>> Cc: 24307@debbugs.gnu.org, Eli Zaretskii <eliz@gnu.org>,
>>   John Wiegley <jwiegley@gmail.com>
>> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
>> Date: Sun, 28 Aug 2016 09:36:55 +0200
>>
>> This morning the keyboard messed up X-wide - backspace in X didn't work
>> any more.
>>
>> Wrote the following into some buffer in sh-mode:
>>
>> ${WERKSTATT=1}
>>
>> See attached png what was displayed, when trying to M-x report-emacs-bug
>> from emacs -Q
> Looks like your Super key is stuck pressed?  Otherwise, why does the
> image show a message about M-s-x being unbound?
>
> Other than that, I see nothing of interest in the image you posted.
> Did I miss something?

Super key was nor pressed nor hang. No idea why that message appeared in 
minibuffer.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Eli Zaretskii]

> Cc: m.kupfer@acm.org, 24307@debbugs.gnu.org, jwiegley@gmail.com
> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
> Date: Mon, 29 Aug 2016 08:09:50 +0200
> 
> >> This morning the keyboard messed up X-wide - backspace in X didn't work
> >> any more.
> >>
> >> Wrote the following into some buffer in sh-mode:
> >>
> >> ${WERKSTATT=1}
> >>
> >> See attached png what was displayed, when trying to M-x report-emacs-bug
> >> from emacs -Q
> > Looks like your Super key is stuck pressed?  Otherwise, why does the
> > image show a message about M-s-x being unbound?
> >
> > Other than that, I see nothing of interest in the image you posted.
> > Did I miss something?
> 
> Super key was nor pressed nor hang. No idea why that message appeared in 
> minibuffer.

However, that's our only clue at this point.  If the OS thought the
Super key was pressed, it could explain why Backspace didn't work, for
example.

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Mike Kupfer]

Eli Zaretskii wrote:

> > Super key was nor pressed nor hang. No idea why that message appeared in 
> > minibuffer.
> 
> However, that's our only clue at this point.  If the OS thought the
> Super key was pressed, it could explain why Backspace didn't work, for
> example.

I noticed that Andreas's Emacs is configured to use ibus.  Could that be
contributing to the problems?

mike

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

On 30.08.2016 00:32, Mike Kupfer wrote:
> Eli Zaretskii wrote:
>
>>> Super key was nor pressed nor hang. No idea why that message appeared in
>>> minibuffer.
>> However, that's our only clue at this point.  If the OS thought the
>> Super key was pressed, it could explain why Backspace didn't work, for
>> example.
> I noticed that Andreas's Emacs is configured to use ibus.  Could that be
> contributing to the problems?
>
> mike

Maybe was at a library that day, which aims to provide unanimous 
key-definitions for related commands in all modes.
https://github.com/andreas-roehler/werkstatt/tree/master/general-key

bug#24307: GNU Emacs 25.1.1 repeated system crash

[Andreas Röhler]

close #24307

As no specific info available, lets close it.

On 25.08.2016 09:01, Andreas Röhler wrote:
> GNU Emacs 25.1.1 (i686-pc-linux-gnu, GTK+ Version 3.14.5) of 2016-08-22
> Linux [...] 3.16.0-4-686-pae #1 SMP Debian 3.16.7-ckt25-2+deb8u3 
> (2016-07-02) i686 GNU/Linux
>
> Last days recent pretest caused two times a system freeze, restart 
> required.
> Unfortunately  couldn't reproduce with Emacs -Q
> As this didn't happen last year --but sometimes the years before-- 
> maybe it's a valuable info though.
> A crash happened after M-x edit-abbrevs RET, editing
>
> "gna"           0    "(< 0 (abs"
> "wgna"           0    "(when (< 0 (abs"
>
>
>
>
>
>