Re: Reports 'Scan error' over and over and can become unresponsive

[Michael Albinus]

Todd Goldfinger <goldfita@gmail.com> writes:

Hi Todd,

> I have been advised to submit a bug report.

... and here I am :-)

> When I set tramp-save-ad-hoc-proxies to true, I see 'Scan error' over
> and over. Depending on my configuration, emacs can become completely
> unresponsive. My setup is a bit unusual, but I've tried this every way
> I can think of: inside docker in WSL2 with my init file inside docker,
> inside docker in WSL2 with my init file on Windows, the same version
> of emacs running only in WSL2 (no docker). I have the same issue no
> matter how I run it.
>
> https://emacs.stackexchange.com/questions/77320/tramp-scan-error
>
> I was able to simplify my init.el down to the following, and I'm still
> having this issue.
>
> (require 'tramp)
> (setq tramp-default-method "ssh")
> (setq tramp-default-host "192.168.2.6")
> (setq tramp-save-ad-hoc-proxies t)
>
>> uname -r
> 5.15.90.1-microsoft-standard-WSL2
>> more /etc/system-release
> Rocky Linux release 9.1 (Blue Onyx)
>
> The docker image is ubuntu:20.04.

Just to complete: you are running

--8<---------------cut here---------------start------------->8---
Emacs  : GNU Emacs 29.0.90 (build 2, x86_64-pc-linux-gnu, GTK+ Version 3.24.20, 
cairo version 1.16.0)
 of 2023-05-11
Package: tramp (2.6.0.29.1 nil/nil)
--8<---------------cut here---------------end--------------->8---

And you have tried to open "/-:|sudo::".

I don't run MS Windows (and WSL), so I'm trying to reproduce on my
Fedora laptop.

I've moved away my ~/.emacs, replacing it by your init.el (except a
different IP address). Then I have called

--8<---------------cut here---------------start------------->8---
# ~/src/emacs-29/src/emacs '/-:|sudo::'
--8<---------------cut here---------------end--------------->8---

No problem, it opens the respective directory in dired-mode. Then I have
exited Emacs, and as expected, the custom settings have appeared in my
.emacs, including the setting for tramp-default-proxies-alist. They look
proper, similar to what you have shown in

> Tramp bug report here
> https://gist.github.com/goldfita/efb3b92d7435c3cac00132bb1f785d5d

Could you, please, try to reproduce my scenario?

And could you please apply (setq debug-on-error t) before you reproduce
the problem? The backtrace should tell us where the scan error comes from.

> Todd

Best regards, Michael.