Bug 4500

pinging once again :)

Bug confirmed with systemd-43-1.mga2.src.rpm

When using sysvinit, the mount is handled by /etc/rc.d/rc.sysinit
which results in
grep usb /proc/self/mounts
/proc/bus/usb /proc/bus/usb usbfs rw,relatime 0 0

When using systemd, /proc/bus/usb is not mounted.

CC: (none) => davidwhodgins

OK, looking into this.

Status: NEW => ASSIGNED
CC: (none) => mageia
Assignee: bugsquad => mageia

Hmm, usbfs is pretty deprecated now I guess...  I suppose the modem driver should be updated.... but that said, I suppose we should create a mount unit for it.

Created attachment 1837 [details]
Systemd patch to mount /proc/bus/usb

This is my proposed patch, but I want to check with upstream that I'm not doing something stupid.

It works here (to do the mount at least), so I have tested it, but not sure if there is some conceptual flaw in my approach.

@colin

I just returned from a weekend on 2G modem (I know I'm with the wrong network ;-)
and all of a sudden I remembered this bug:

it seems not so much a problem of the option driver but more a problem of the drakxtools...

I started digging a bit deeper into this matter and found out the following:

* once you say controlled by NM: no, things work already much better ;-)
* once you mount usbfs draknetwork shows the modem 
* I must have been wrong with option driver not attached to the modem, as it is shown in dmesg as providing 2 ttys, quite early in the boot process
* while playing with AT commands I got the impression the modem is there while it's not shown in draknet


now back home I tried the following:
ifup ppp0

and here I'm writing this comment with:
[root@localhost diego]# mount |grep usb
[root@localhost diego]# 

sorry for misleading you

OK, so that makes sense that it's really needed for draknet. So really we need to migrate it to not use usbfs.

If that's all, then I think my proposed patch will do for now and we can refactor for mga3 to avoid it.

i dont care why it works as long as it does ;-)

is there already a tracker for 3?

we can open one for the specifications (an this bug needs a new summary then imho)

we might as well close it as it is fixed for 2 and reopen a new one with the proper description for 3?

OK, so I've applied my patch for now. It should be revisited later however.

Status: ASSIGNED => RESOLVED
Resolution: (none) => FIXED

Since the latest kernel updates in Cauldron - I am no longer able to see
my WAN modem again (Ericsson F5521gw)

The solution below doesn't work either . . . 

[root@ThinkFox rfox]# lsmod | grep cdc
cdc_ncm                18345  0 
usbnet                 36447  1 cdc_ncm
cdc_wdm                18885  0 
cdc_acm                28752  0 
usbcore               212051  9
btusb,uvcvideo,ehci_hcd,usbhid,usbnet,cdc_acm,cdc_ncm,cdc_wdm,xhci_hcd
[root@ThinkFox rfox]# mount -n -t usbfs -o
devmode=0664,devgid=43 /proc/bus/usb /proc/bus/usb
mount: mount point /proc/bus/usb does not exist

Looks like /proc/bus/ doesn't have usb anymore:
[root@ThinkFox rfox]# ls /proc/bus/
input/ pci/

Should I raise a bug?

Thx,
R.Fox

On Wed, 2012-03-28 at 09:15 +0100, Colin Guthrie wrote:
> 'Twas brillig, and Robert Fox at 27/03/12 20:11 did gyre and gimble:
> > Since recent Cauldron updates, I can't get my 3G card (Ericsson F5521gw)
> > to work - looks like the modules are loading:
> > 
> > [root@ThinkFox rfox]# lsmod | grep cdc
> > cdc_ncm                18343  0 
> > usbnet                 32360  1 cdc_ncm
> > cdc_acm                28712  0 
> > cdc_wdm                18241  0 
> > usbcore               206950  10
> > uvcvideo,btusb,cdc_ncm,usbnet,cdc_acm,cdc_wdm,xhci_hcd,usbhid,ehci_hcd
> > 
> > but when I start drakconnect, I get "no device for this connection type"
> > 
> > On the same notebook (Thinkpad x220) it works fine under Ubuntu 12.04
> > B1)
> > 
> > Any hints would be welcome.
> 
> Try running as root:
> 
> mount -n -t usbfs -o devmode=0664,devgid=43 /proc/bus/usb /proc/bus/usb
> 
> 
> Then run drakconnect.
> 
> aka: https://bugs.mageia.org/show_bug.cgi?id=4500
> 
> Col
> 

dmesg | grep cdc
[   12.652393] cdc_wdm 3-1.4:1.5: cdc-wdm0: USB WDM device
[   12.652450] cdc_wdm 3-1.4:1.8: cdc-wdm1: USB WDM device
[   12.652466] usbcore: registered new interface driver cdc_wdm
[   12.654471] cdc_acm 3-1.4:1.1: ttyACM0: USB ACM device
[   12.655552] cdc_acm 3-1.4:1.3: ttyACM1: USB ACM device
[   12.660486] cdc_acm 3-1.4:1.9: ttyACM2: USB ACM device
[   12.668410] usbcore: registered new interface driver cdc_acm
[   12.668412] cdc_acm: USB Abstract Control Model driver for USB modems and
ISDN adapters
[   12.689107] cdc_ncm 3-1.4:1.6: usb0: register 'cdc_ncm' at
usb-0000:00:1d.0-1.4, CDC NCM, 02:80:37:ec:02:00
[   12.689131] usbcore: registered new interface driver cdc_ncm
[   16.497080] cdc_ncm: usb0: network connection: disconnected

	

    Collapse All Comments
    Expand All Comments

Status: RESOLVED => REOPENED
CC: (none) => rfox
Resolution: FIXED => (none)

Please open a new bug rather than reopening an old one that is linked to trackers etc. It's 99% likely is a different cause even if the end result is the same. This fix is also likely just to port whatever needs /proc/bus/usb to use newer APIs anyway, so likely best to open bug reports against those things directly.

Status: REOPENED => RESOLVED
Resolution: (none) => FIXED

Have opened a new bug:

https://bugs.mageia.org/show_bug.cgi?id=7039

Thx -