bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

To: bug-gnu-emacs@gnu.org
Subject: 24.5; emacs unusable in minibuffer, unpredictably, with "k is
undefined" for every key
--text follows this line--
I have long had a bug where everything I type says "k is undefined",
where k stands for any key including ESC and C-g.  this occurs
randomly.

It is impossible to get out of this mode, unless you can figure out
some SIGUSR1 voodoo.  Except once, I think type-break activating got
me out of the mode.  And once I got
out by just coming back to it after a half hour or something.

I suspect maybe type-break also causes the bug, if certain conditions
obtain.  I am not sure.  But it is the only timer-based thing I can
think of.

I believe it occurs when I am in the minibuffer.  Perhaps only at a
password prompt.

I don't know if this bug report is useful.

Thanks.

In GNU Emacs 24.5.1 (x86_64-pc-linux-gnu, GTK+ Version 3.22.11)
 of 2017-09-12 on hullmann, modified by Debian
Windowing system distributor `The X.Org Foundation', version 11.0.11902000
System Description:	Debian GNU/Linux 9.11 (stretch)

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Eli Zaretskii]

> From: Samuel Wales <samologist@gmail.com>
> Date: Sun, 29 Dec 2019 18:20:15 -0700
> 
> I have long had a bug where everything I type says "k is undefined",
> where k stands for any key including ESC and C-g.  this occurs
> randomly.
> 
> It is impossible to get out of this mode, unless you can figure out
> some SIGUSR1 voodoo.  Except once, I think type-break activating got
> me out of the mode.  And once I got
> out by just coming back to it after a half hour or something.

Can you show a screenshot of how your Emacs session looks when this
happens?

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

On 12/30/19, Eli Zaretskii <eliz@gnu.org> wrote:
> Can you show a screenshot of how your Emacs session looks when this
> happens?

it is quite infrequent.  basically, last time, the screen showed the
buffer, password prompt in minibuffer, and then whenever you type any
key, it says that key is undefined.

-- 
The Kafka Pandemic

What is misopathy?
https://thekafkapandemic.blogspot.com/2013/10/why-some-diseases-are-wronged.html

The disease DOES progress. MANY people have died from it. And ANYBODY
can get it at any time.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Eli Zaretskii]

> From: Samuel Wales <samologist@gmail.com>
> Date: Tue, 31 Dec 2019 15:39:36 -0700
> Cc: 38802@debbugs.gnu.org
> 
> On 12/30/19, Eli Zaretskii <eliz@gnu.org> wrote:
> > Can you show a screenshot of how your Emacs session looks when this
> > happens?
> 
> it is quite infrequent.  basically, last time, the screen showed the
> buffer, password prompt in minibuffer, and then whenever you type any
> key, it says that key is undefined.

Please post a screenshot next time it happens.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Michael Heerdegen]

Eli Zaretskii <eliz@gnu.org> writes:

> Can you show a screenshot of how your Emacs session looks when this
> happens?

Dunno if it's related, but I sometimes saw something similar when I had
been kicked out of an y-or-n-p prompt somehow, AFAIR.  But I'm currently
not able to provoke something like this.

Michael.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Eli Zaretskii]

> From: Michael Heerdegen <michael_heerdegen@web.de>
> Cc: Samuel Wales <samologist@gmail.com>,  38802@debbugs.gnu.org
> Date: Wed, 01 Jan 2020 01:31:01 +0100
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Can you show a screenshot of how your Emacs session looks when this
> > happens?
> 
> Dunno if it's related, but I sometimes saw something similar when I had
> been kicked out of an y-or-n-p prompt somehow, AFAIR.  But I'm currently
> not able to provoke something like this.

You are saying that sometimes y-or-n-p doesn't react to C-g?

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Michael Heerdegen]

Eli Zaretskii <eliz@gnu.org> writes:

> You are saying that sometimes y-or-n-p doesn't react to C-g?

No, when that happens, the y-or-n-p prompt is gone.  Everything looks
like normal top level, but all keys are "undefined" (maybe all except
"y" and "n"?).

I can't be more specific, haven't seen the problem for a long time now.
AFAIR that happened when I somehow stepped out of the prompt, e.g. by
forcing Emacs into debugger, or using Edebug.  I don't recall exactly,
and maybe it is unrelated, but there are similarities to what the OP
described.  Too bad I currently can't reproduce the issue I mean.  I
think the OP can be of more help.

Michael.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

i'd be delighted to be of more help.  however, besides posting a
screenshot next time it occurs [which is infrequent enough that it
could be /quite/ a long time], what would you suggest i do to be of
more help?

i have no idea why you said that.  i thought it could be useful to
somebody to post the bug report, in case it jogs somebody's memory of
code interaction, although there wasn't much to it.  i don't want to
regret doing so.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

i will try to post a screenshot next time.  it could be quite a long time.

i think i forgot to mention that the password prompt, in the most
recent instance, was triggered by shell mode.

any suggestions of what to put in for a sigusr1 or sigusr2  would be
interesting.  i could put debug there maybe.  or find out what modes
are active, or the values of hooks.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

i think i misunderstood "can be of more help" to mean "is not helping"
instead of "can be of more help than me".  i get it now :).

although there are few things i can do.  :)

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Michael Heerdegen]

Samuel Wales <samologist@gmail.com> writes:

> i think i misunderstood "can be of more help" to mean "is not helping"
> instead of "can be of more help than me".  i get it now :).

Indeed - and I'm very sorry for the unintentional insult :-(  I guess I
really have to improve my English until something really bad happens one
day...

Michael.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

your english is very good and your politeness impeccable.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Eli Zaretskii]

> From: Samuel Wales <samologist@gmail.com>
> Date: Tue, 31 Dec 2019 22:39:37 -0700
> Cc: Eli Zaretskii <eliz@gnu.org>, 38802@debbugs.gnu.org
> 
> i will try to post a screenshot next time.  it could be quite a long time.

That's OK.  If this problem happens to others, they might come up with
data sooner; if not, it means the problem is rare enough for it not to
be urgent.

Btw, if you can afford that, I'd suggest to upgrade to a newer version
of Emacs, because perhaps this problem was already solved.  The
current version is 26.3, and 27.1 will start its pretest soon.

> any suggestions of what to put in for a sigusr1 or sigusr2  would be
> interesting.

I'd suggest to use debug-on-event.  You can read about it in the ELisp
manual, node "Error Debugging".

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

it is quite rare, but has been around for many emacs versions.

i could test my intuition that it might be related to type-break by
setting type-break to trigger frequently, then running passwd in the
shell, but i don't have it in me atm.

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Lars Ingebrigtsen]

Samuel Wales <samologist@gmail.com> writes:

> it is quite rare, but has been around for many emacs versions.
>
> i could test my intuition that it might be related to type-break by
> setting type-break to trigger frequently, then running passwd in the
> shell, but i don't have it in me atm.

This was some months ago.  Have you seen this problem again?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Lars Ingebrigtsen]

Lars Ingebrigtsen <larsi@gnus.org> writes:

> Samuel Wales <samologist@gmail.com> writes:
>
>> it is quite rare, but has been around for many emacs versions.
>>
>> i could test my intuition that it might be related to type-break by
>> setting type-break to trigger frequently, then running passwd in the
>> shell, but i don't have it in me atm.
>
> This was some months ago.  Have you seen this problem again?

More information was requested one month ago, but was not given, so
I'm closing this bug report.  If further progress can be made here,
please respond to the debbugs address and we'll reopen.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Samuel Wales]

have not seen it since last time.  it is pretty rare.

if you are in the middle of some critical irc conversation, have a lot
of unsaved buffers, etc.  it can feel like you've been denied the
source and life of all computing -- emacs.

if it needs closing, then whatever is appropriate.

for health reason i am limited in computer use.

On 9/22/20, Lars Ingebrigtsen <larsi@gnus.org> wrote:
> Samuel Wales <samologist@gmail.com> writes:
>
>> it is quite rare, but has been around for many emacs versions.
>>
>> i could test my intuition that it might be related to type-break by
>> setting type-break to trigger frequently, then running passwd in the
>> shell, but i don't have it in me atm.
>
> This was some months ago.  Have you seen this problem again?
>
> --
> (domestic pets only, the antidote for overdose, milk.)
>    bloggy blog: http://lars.ingebrigtsen.no
>


-- 
The Kafka Pandemic

Please learn what misopathy is.
https://thekafkapandemic.blogspot.com/2013/10/why-some-diseases-are-wronged.html

bug#38802: 24.5; emacs unusable, unpredictably, with "k is undefined" for every key

[Lars Ingebrigtsen]

Samuel Wales <samologist@gmail.com> writes:

> have not seen it since last time.  it is pretty rare.
>
> if you are in the middle of some critical irc conversation, have a lot
> of unsaved buffers, etc.  it can feel like you've been denied the
> source and life of all computing -- emacs.
>
> if it needs closing, then whatever is appropriate.

OK, it doesn't sound likely that we'll make further progress in this bug
report, so I'm closing it.  If you come up with a recipe for reproducing
the bug, please respond to the debbugs address and we'll reopen.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no