Devuan bug report logs -
#333
usb-modeswitch no longer creates /dev/gsmmodem symlink
Reply or subscribe to this bug.
Toggle useless messages
Report forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#333
; Package usb-modeswitch-data
.
(full text, mbox, link).
Acknowledgement sent to Guenther Brunthaler <gb_about_gnu@gmx.net>
:
New bug report received and forwarded. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #5 received at submit@bugs.devuan.org (full text, mbox, reply):
Package: usb-modeswitch-data
Version: 20170120-1
The /dev/gsmmodem symlink is very useful as the device specified in
/etc/ppp/peers/* configurations, because it works no matter what modem
has actually been connected to the computer.
The eudev-rule from the the beforementioned package invokes the
usb_modeswitch binary with the --symlink-name which used to create the
/dev/gsmmodem symlink:
$ grep -A2 gsmmodem /lib/udev/rules.d/40-usb_modeswitch.rules
# Adds a symlink "gsmmodem[n]" to the lowest ttyUSB port with interrupt
# transfer; checked against a list of known modems, or else no action
KERNEL=="ttyUSB*", ATTRS{bNumConfigurations}=="*",
PROGRAM="usb_modeswitch --symlink-name %p %s{idVendor} %s{idProduct}
%E{PRODUCT}", SYMLINK+="%c"
The binary ignores the option, and hence no symlink is created.
I think the udev rule is incorrect and should invoke the
/lib/udev/usb_modeswitch shell script rather than the
/usr/sbin/usb_modeswitch binary, because only the shell script supports
the --symlink-name option.
This is at least the case in the current version 2.5.0+repack0-1 of the
usb-modeswitch package.
I am using Devuan GNU/Linux 2.0 (ascii)
on branch stretch
with kernel
Linux xeeepc 4.9.0-9-686 #1 SMP Debian 4.9.168-1+deb9u2 (2019-05-13)
i686 GNU/Linux
Information forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#333
; Package usb-modeswitch-data
.
(full text, mbox, link).
Acknowledgement sent to Guenther Brunthaler <gb_about_gnu@gmx.net>
:
Extra info received and forwarded to list. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #10 received at 333@bugs.devuan.org (full text, mbox, reply):
First of all, I am using the "ascii" branch, not "stretch".
I also think now the problem is not the rule-file but rather the fact
that the PATH is set up wrong for the eudev service.
The following modification made the problem go away:
$ diff -u /etc/udev/udev.conf.upstream /etc/udev/udev.conf
--- /etc/udev/udev.conf.upstream 2019-06-13 23:14:29.000000000 +0200
+++ /etc/udev/udev.conf 2019-06-13 23:15:17.000000000 +0200
@@ -4,3 +4,5 @@
# initramfs should be rebuilt.
#udev_log="info"
+
+PATH=/lib/udev:$PATH
because it puts /lib/udev at the front of the PATH, making the udev rule
invoke the correct executable.
Information forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#333
; Package usb-modeswitch-data
.
(full text, mbox, link).
Acknowledgement sent to Guenther Brunthaler <gb_about_gnu@gmx.net>
:
Extra info received and forwarded to list. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #15 received at 333@bugs.devuan.org (full text, mbox, reply):
I have just verified that upstream uses the same udev rule file:
https://salsa.debian.org/debian/usb-modeswitch-data/raw/debian/master/40-usb_modeswitch.rules
Therefore upstream obviously expectes /lib/udev to be part of the $PATH,
meaning that it cannot be usb-modeswitch-data's fault if this is not the
case.
Created a new bug entry for the eudev package.
Please close this bug!
Reply sent
to svante.signell@gmail.com
:
You have taken responsibility.
(Tue, 14 Jan 2020 13:12:23 GMT) (full text, mbox, link).
Notification sent
to Guenther Brunthaler <gb_about_gnu@gmx.net>
:
bug acknowledged by developer.
(Tue, 14 Jan 2020 13:12:25 GMT) (full text, mbox, link).
Message #20 received at 333-close@bugs.devuan.org (full text, mbox, reply):
Send a report that this bug log contains spam.