On Thu, 2010-10-14 at 09:30 +0200, van Schelve wrote: > Am Mittwoch, den 13.10.2010, 21:00 +0200 schrieb van Schelve > <[email protected]>: > > Am Mittwoch, den 13.10.2010, 12:49 +0200 schrieb Dan Williams > > <[email protected]>: > >> On Mon, 2010-10-11 at 10:43 +0200, van Schelve wrote: > >>> Hi, > >>> > >>> I found a different problem where I get reports from our users in the > >>> field. They tell us that they are not able to connect to mobile > >>> network. > >>> > >>> In the logs I captured I found a cme error 30 which means "No network > >>> service". I tried it two or three times with still the same problem. > >>> Then I tested with a second identical laptop with an identical setup > >>> and I was able to bring up the mobile connection on this system without > >>> any problem. > >>> > >>> So I decided to use the rfkill switch on the laptop where I found the > >>> problem with the result that dialin was possible after doing th rfkill > >>> cycle. > >> > >> Interesting; your module seems to have firmware 1.1.8.3, while I have 3 > >> different modules with 1.1.8.15, 1.1.9.3, and 2.0.8.19. I think > >> 2.0.8.19 screwed my device though, so don't try it. I've noticed odd > >> problems like this on other Sierra devices in the past that were solved > >> by firmware updates (AC875 and AC881), but it's pretty hard to find > >> firmware upgrades for their embedded modules. Do you know offhand if > >> any of the devices in the field has other firmware than 1.1.8.3? > > > > We have completely identical hardware in the field. All the systems are > > Lenovo Thinkpad T61 with identical BIOS 2.10 version and the same > > firmware > > on the embedded sierra device : 1.1.8.3. I think changing the firmware > > to > > a different version woud be a really bad option. Our systems are spread > > all over germany and I don't think the firmware update would be > > possible > > under linux. > > I talked about this problem with a contact at SierraWireless. He said > that > a firmware upgrade will not help in this case. But what can help is > checking > for PS attach and dial only if PS attach status is attached. Dialing > without > PS attach can bring modem into unwanted condition that sometimes > restart is > needed to recover.
Hmm, interesting. I'll try a few things out then. Dan > > > > On the other hand it was good luck that I was able to see this issue on > > my system. I'm thinking about enabling debug on all our systems this > > way: > > > > cat > > /usr/share/dbus-1/system-services/org.freedesktop.ModemManager.service > > [D-BUS Service] > > Name=org.freedesktop.ModemManager > > Exec=/usr/sbin/modem-manager --debug >> > > /var/log/modem-manager-debug.log > > User=root > > > > and starting NetworkManager --log-level=debug in the upstart initscript > > > > What do you mean? Would this be ok? > > > >> > >> In any case, I'll add some code to ModemManager to get better Sierra > >> device status for GSM devices to help debug this issue. If you run into > >> this problem yourself, we want the output of AT!GSTATUS? to get more > >> info if you can. > >> > > > > But this means I need to disable mobile broadband temporary because > > otherwise > > the serial port is blocked by modem-manager and I cannot get the > > information at > > the same time. Having more debug in sierraplugin would be very helpful > > because > > I have seen other issues that I have not reported yet (for example SIM > > Busy) > > > > HG > > > >> Dan > _______________________________________________ networkmanager-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/networkmanager-list
