You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Yves-Alexis Perez reported that since 2.x, cups clients don't connect to the correct socket: https://bugs.debian.org/792730.
I've tried to debug this, and my analysis is that since 2.x, lpstat -H reports localhost:631 instead of /var/run/cups/cups.sock , even with the socket file as follows in both 1.7 and 2.0:
Description: In absence of client.conf configuration, the default domain socket
(/var/run/cups/cups.sock) will get checked for read, write and execution bits,
but the latter doesn't make sense
Author: Didier Raboud odyx@debian.org
Origin: vendor
Bug-Debian: https://bugs.debian.org/792730
Last-Update: 2015-07-18
--- a/cups/usersys.c
+++ b/cups/usersys.c
@@ -1017,7 +1017,7 @@
struct stat sockinfo; /* Domain socket information */
if (!stat(CUPS_DEFAULT_DOMAINSOCKET, &sockinfo) &&
Version: 2.1-current
CUPS.org User: odyx
Yves-Alexis Perez reported that since 2.x, cups clients don't connect to the correct socket: https://bugs.debian.org/792730.
I've tried to debug this, and my analysis is that since 2.x,
lpstat -H
reports localhost:631 instead of /var/run/cups/cups.sock , even with the socket file as follows in both 1.7 and 2.0:srw-rw-rw- 1 root root 0 jui 18 12:36 /var/run/cups/cups.sock
I've traced this to be correctible by changing the access checks in cups_finalize_client_conf, with the attached patch.
Does this make sense?
The text was updated successfully, but these errors were encountered: