Bug 1352687

Summary:	Some Huawei hostless modems not switched to modem mode automatically (udev-rules)
Product:	[Fedora] Fedora	Reporter:	samoht0 <samoht0-bugzilla>
Component:	usb_modeswitch	Assignee:	romal <redhat>
Status:	CLOSED WORKSFORME	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	24	CC:	huzaifas, lkundrak, peter.kovar, redhat
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2016-07-10 13:28:30 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description samoht0 2016-07-04 17:58:33 UTC

Description of problem:

I've got a Huawei E3372 hostless modem, which worked fine in F22. With F24 it's stuck in Storage-Mode after plugging in. No udev/systemd activity shown in system-journal. Switching manually via usb_modeswitch works. Matching config-data is present.

So my guess is to blame the usb_modeswitch udev-rules.

Likely the same issue here (E3372 + E1800):

https://bugs.launchpad.net/ubuntu/+source/usb-modeswitch/+bug/1562431

Version-Release number of selected component (if applicable):

usb_modeswitch-2.3.0-1.fc24

Additional info:

My old "classic" broadband modem Huawei E169 works fine.

There is a patch included in usb_modeswitch 2.4.0, which improves systemd-integration. The commit and the forum post, which are linked in the changelog, look like a similar issue.

http://www.draisberghof.de/usb_modeswitch/ChangeLog

So suggestion is to try 2.4.0 in F24 testing (probably without karma based stable).

Comment 1 Peter Kovář 2016-07-09 09:01:57 UTC

/usr/lib/udev/rules.d/40-usb_modeswitch.rules

# Huawei E3372
ATTR{idVendor}=="12d1", ATTR{idProduct}=="14fe", RUN+="usb_modeswitch '%b/%k'"

Comment 2 samoht0 2016-07-09 09:53:29 UTC

To be precise: Yes, the specific rule is present and worked fine in F22 with systemd 219.

But there is a regression with later systemd versions as shipped with F24. This is analyzed and confirmed by the developer in the upstream forum post (based on Archlinux with systemd 221):

http://www.draisberghof.de/usb_modeswitch/bb/viewtopic.php?p=16777

/usr/lib/udev/usb_modeswitch was changed in 2.4.0 to fix this. So my suggestion was, that we could try 2.4.0, which is supposed to fix the regression.

Comment 3 samoht0 2016-07-10 09:50:07 UTC

I tried 2.4.0 from F25, but sadly no change in behavior. So my bug is most likely another issue.

Comment 4 samoht0 2016-07-10 13:28:30 UTC

(In reply to Peter Kovář from comment #1)
> /usr/lib/udev/rules.d/40-usb_modeswitch.rules
> 
> # Huawei E3372
> ATTR{idVendor}=="12d1", ATTR{idProduct}=="14fe", RUN+="usb_modeswitch
> '%b/%k'"

Now it's solved. Your advice was good, but I had to change the Id (my fault, as I have the correct Id in systemd-journal):

# Huawei E3372
ATTR{idVendor}=="12d1", ATTR{idProduct}=="1f01", RUN+="usb_modeswitch '%b/%k'"

I wonder why this worked out of box in F22...