Bug 19523

Summary: Huawei dongle no longer seen by drakx-net
Product: Mageia Reporter: Marja Van Waes <marja11>
Component: RPM PackagesAssignee: Mageia Bug Squad <bugsquad>
Status: RESOLVED FIXED QA Contact:
Severity: normal    
Priority: Normal CC: geiger.david68210, ghibomgx, mageiatools, thierry.vignaud
Version: Cauldron   
Target Milestone: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Source RPM: usb_modeswitch-2.4.0-1.mga6, usb_modeswitch-data-20160803-1.mga6 CVE:
Status comment:

Description Marja Van Waes 2016-10-04 13:04:29 CEST
In yesterday-last-updated cauldron, my 3G dongle no longer works.
It still works in (not recently updated) Mageia 5


[marja@cldrn_64 ~]$ lspcidrake -v | grep -i huawei
usb_storage     : HUAWEI|HUAWEI Mobile [Mass Storage|SCSI|Bulk-Only] (vendor:12d1 device:14fe)
[marja@cldrn_64 ~]$ lsusb | grep -i huawei
Bus 002 Device 004: ID 12d1:14fe Huawei Technologies Co., Ltd. Modem (Mass Storage Mode)
[marja@cldrn_64 ~]$ 

I'll reboot into Mga5 to get the output of the above commands before updating (_if_ I update :-þ)
Comment 1 Thierry Vignaud 2016-10-04 13:18:30 CEST
Have you installed usb_modeswitch?

Keywords: (none) => NEEDINFO
CC: (none) => thierry.vignaud

Comment 2 Marja Van Waes 2016-10-04 13:27:07 CEST
(In reply to Thierry Vignaud from comment #1)
> Have you installed usb_modeswitch?

I'll check after rebooting

Just for the record:

In Mga5 there is no problem with (probably old):

usb_modeswitch-2.2.1-1.mga5
lib64ldetect0.12-0.12.8-1.mga5
ldetect-0.12.8-1.mga5
ldetect-lst-0.1.346.4-1.mga5

[marja@Mga5_64bit ~]$ lspcidrake -v | grep -i huawei
option          : HUAWEI|HUAWEI Mobile (vendor:12d1 device:1506)
[marja@Mga5_64bit ~]$ lsusb | grep -i huawei
Bus 002 Device 005: ID 12d1:1506 Huawei Technologies Co., Ltd. Modem/Networkcard
[marja@Mga5_64bit ~]$
Comment 3 Marja Van Waes 2016-10-04 13:35:19 CEST
(In reply to Thierry Vignaud from comment #1)
> Have you installed usb_modeswitch?

Yes

[marja@cldrn_64 ~]$ rpm -qa | grep usb_modeswitch
usb_modeswitch-2.4.0-1.mga6
usb_modeswitch-data-20160803-1.mga6
[marja@cldrn_64 ~]$

Keywords: NEEDINFO => (none)

Comment 4 Thierry Vignaud 2016-10-04 13:51:21 CEST
Did you try running usb_modeswitch as root?
Comment 5 Marja Van Waes 2016-10-04 14:09:41 CEST
(In reply to Thierry Vignaud from comment #4)
> Did you try running usb_modeswitch as root?

Sorry, I'm googling myself silly, but I don't see how to give a switching method


# usb_modeswitch -v 12d1 -p 14fe
Look for default devices ...
   product ID matched
 Found devices in default mode (1)
Access device 005 on bus 002
Get the current device configuration ...
Current configuration number is 1
Use interface number 0

USB description data (for identification)
-------------------------
Manufacturer: HUAWEI
     Product: HUAWEI Mobile
  Serial No.: not provided
-------------------------
Warning: no switching method given. See documentation
-> Run lsusb to note any changes. Bye!
Comment 6 Marja Van Waes 2016-10-04 14:13:20 CEST
(well, I did see some '-M "a_very_long_string"' in examples, but I don't dare to blindly try that)
Thierry Vignaud 2016-10-04 14:51:39 CEST

Source RPM: (none) => usb_modeswitch?

Comment 7 Marja Van Waes 2016-10-04 15:02:19 CEST

CC'ing usb_modeswitch maintainer, to help find out whether this is a usb_modeswitch issue.

Btw, if it is:
Is the kernel maintainer group only meant for packages with "kernel" in the name?
usb_modeswitch is in the "System/Kernel and hardware" group

CC: (none) => geiger.david68210

Comment 8 David GEIGER 2016-10-04 15:06:50 CEST
Hmmm! This is why I did not do this update for these packages but someone else has done this update in my place before asking me :(
Comment 9 Marja Van Waes 2016-10-04 15:20:10 CEST
(In reply to David GEIGER from comment #8)
> Hmmm! This is why I did not do this update for these packages but someone
> else has done this update in my place before asking me :(

I'll try to find a copy of 

   usb_modeswitch-data-20160112-1.mga6 and 
   usb_modeswitch-2.3.0-1.mga6

tonight and downgrade to them.

CC: (none) => ghibomgx
Source RPM: usb_modeswitch? => usb_modeswitch-2.4.0-1.mga6, usb_modeswitch-data-20160803-1.mga6

Comment 10 David GEIGER 2016-10-04 15:30:07 CEST
Maybe you can try fedora's packages to see what's happens:

http://koji.fedoraproject.org/koji/buildinfo?buildID=785537
http://koji.fedoraproject.org/koji/buildinfo?buildID=785540
Comment 11 Thierry Vignaud 2016-10-04 15:56:37 CEST
Or try usb_modeswitch-2.4.0-2.mga6
I've added a missing systemd unit + 2 fixes from FC (a crash fix and an undefined memory issue)
Comment 12 Thierry Vignaud 2016-10-04 15:58:19 CEST
(In reply to David GEIGER from comment #8)
Well the thing is, it's needed for supporting some devices (see http://www.draisberghof.de/usb_modeswitch/ChangeLogData)
I guess that's why it got updated
Comment 13 Marja Van Waes 2016-10-04 21:48:51 CEST
(In reply to Thierry Vignaud from comment #11)
> Or try usb_modeswitch-2.4.0-2.mga6
> I've added a missing systemd unit + 2 fixes from FC (a crash fix and an
> undefined memory issue)

Thanks Thierry :-)

Now lsusb shows:

Bus 002 Device 005: ID 12d1:1506 Huawei Technologies Co., Ltd. Modem/Networkcard

And internet works :-)

I'll open another report tomorrow for a remaining issue: draknetcenter does not show I'm connected. 
It shows HUAWEI HUAWEI Mobile as Wired (Ethernet) wwp0s29f7u1i1
But when checking the Network Monitor, there is _no_ trafic for wwp0s29f7u1i1,
but there _is_ trafic in a the tab with "mobile or dongle icon"ppp0

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

Comment 14 Marja Van Waes 2016-10-05 11:23:29 CEST
(In reply to Marja van Waes from comment #13)

> 
> I'll open another report tomorrow for a remaining issue: draknetcenter does
> not show I'm connected. 
> It shows HUAWEI HUAWEI Mobile as Wired (Ethernet) wwp0s29f7u1i1
> But when checking the Network Monitor, there is _no_ trafic for
> wwp0s29f7u1i1,
> but there _is_ trafic in a the tab with "mobile or dongle icon"ppp0

I'll first try to reproduce this in a different cauldron, I've just spent an hour trying to get good information from my logs etc, but there are too many elements that might make this a corner case.