[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: breakage in emacsclient
From: |
jbranso |
Subject: |
Re: breakage in emacsclient |
Date: |
Sat, 24 Oct 2020 23:23:05 +0000 |
I didn't even know that the latest version of Emacs compiled on the Hurd...
I wish I could be more helpful...How is geeEmacs?
October 24, 2020 7:00 PM, "Andrew Eggenberger" <andrew.eggenberger@gmail.com>
wrote:
> I recently built gccemacs 28 and found that when I tried to connect to
> the server with emacsclient I would encounter the following error:
>
> /home/demo/.local/bin/emacsclient: Invalid socket owner
> /home/demo/.local/bin/emacsclient: No socket or alternate editor.
> Please use:
>
> --socket-name server-file (or environment variable
> --EMACS_SERVER_FILE) alternate-editor (or environment variable
> --ALTERNATE_EDITOR)
>
> After building the vanilla emacs 28 and finding the same issue, I
> looked at the source code of emacsclient.c and found that the uid
> validation in the function set_local_socket since 26.3 (the latest
> version available for hurd on debian). If the uid returned by
> geteuid() doesn't match what fstat puts in the connect_stat.st_uid
> when given the file descriptor for the server, the "Invalid socket
> owner error is triggered." Using gdb I found uid to be 1000 and
> connect_stat.st_uid to be 0.
>
> I filed an emacs bug and it appears to only affect the hurd. Are
> geteuid() and fstat() returning the values we expect?
>
> Andrew