GNOME Bugzilla – Bug 591047
Huawei E1550 3G USB modem (HSDPA USB Stick) cannot connect
Last modified: 2012-07-24 17:13:25 UTC
Please describe the problem: This modem is registered OK and properly configured. But when trying to connect, it disconnects immediately. Steps to reproduce: 1. Connect the modem 2. Set up an account for Oister.dk (now in the Gnome provider list SVN) 3. Restart NetworkManager in debugging mode and try to connect Actual results: The modem disconnects Expected results: The modem should connect Does this happen every time? Yes Other information: Hi, I'm having almost the same problem as Bug 565177 - Huawei E160G 3G USB modem cannot connect (NetworkManager) ( http://bugzilla.gnome.org/show_bug.cgi?id=565177 ), as I am also on the same network, except this time it is about Huawei E1550 modem. First of all, in Ubuntu 9.04 this stick still appears as a USB mass storage drive (or rather, auto-install CD for Windows drivers), so first I did the udev-extras trick (given on "Need hints on adding a new device to Network Manager" http://ubuntuforums.org/showpost.php?p=7622263&postcount=3) mentioned in "Huawei E1550 settings - Ubuntu Forums" - http://ubuntuforums.org/showthread.php?t=1193355. Before the trick, I think I had the same vendor=0x12d1 product=0x1446 as in the post; after the trick (and reinserting the stick), lsusb reports ID 12d1:1001 Huawei Technologies Co., Ltd. E620 USB Modem Then I get the wizard "New Mobile Broadband Connection", which completes fine (my service provider is in the list), it says "New Configuration Created", and when I try to then click on the connection, I get asked for the PIN - the right PIN seems to get accepted, and immediately I get - "GSM Network; Disconnected". On subsequent tries to connect via NetworkManager again, I don't get asked for PIN anymore, but the same "disconnected" message occurs. I tried to restart NetworkManager in debug mode using: sudo NM_SERIAL_DEBUG=1 NM_PPP_DEBUG=1 /usr/sbin/NetworkManager restart and then observe tail -f /var/log/syslog, trace upon attempting to connect is: Aug 7 11:57:29 mycomputer NetworkManager: <info> Activation (ttyUSB3) starting connection 'OiSTER' Aug 7 11:57:29 mycomputer NetworkManager: <info> (ttyUSB3): device state change: 3 -> 4 Aug 7 11:57:29 mycomputer NetworkManager: <info> Activation (ttyUSB3) Stage 1 of 5 (Device Prepare) scheduled... Aug 7 11:57:29 mycomputer NetworkManager: <info> Activation (ttyUSB3) Stage 1 of 5 (Device Prepare) started... Aug 7 11:57:29 mycomputer NetworkManager: <debug> [1249639049.992936] nm_serial_device_open(): (ttyUSB3) opening device... Aug 7 11:57:30 mycomputer NetworkManager: <info> Activation (ttyUSB3) Stage 1 of 5 (Device Prepare) complete. Aug 7 11:57:30 mycomputer NetworkManager: <info> (ttyUSB3): powering up... Aug 7 11:57:30 mycomputer NetworkManager: <WARN> automatic_registration_response(): Automatic registration failed: registration denied. Aug 7 11:57:30 mycomputer NetworkManager: <info> (ttyUSB3): device state change: 4 -> 9 Aug 7 11:57:30 mycomputer NetworkManager: <debug> [1249639050.295291] nm_serial_device_close(): Closing device 'ttyUSB3' Aug 7 11:57:30 mycomputer NetworkManager: <info> Marking connection 'OiSTER' invalid. Aug 7 11:57:30 mycomputer NetworkManager: <info> Activation (ttyUSB3) failed. Aug 7 11:57:30 mycomputer NetworkManager: <info> (ttyUSB3): device state change: 9 -> 3 Aug 7 11:57:30 mycomputer NetworkManager: <info> (ttyUSB3): deactivating device (reason: 0). Aug 7 11:57:30 mycomputer NetworkManager: <WARN> add_ip4_route_to_gateway(): (wlan1): failed to add IPv4 route to gateway (-100) Aug 7 11:57:30 mycomputer NetworkManager: <info> Policy set 'Auto chaozznet' (wlan1) as default for routing and DNS. Aug 7 11:57:30 mycomputer NetworkManager: nm_system_device_flush_ip4_routes_with_iface: assertion `iface_idx >= 0' failed Aug 7 11:57:30 mycomputer NetworkManager: nm_system_device_flush_ip4_addresses_with_iface: assertion `iface_idx >= 0' failed Also to finally mention that after this, I tried wvdial as recommended by the service provider (https://www.oister.dk/selfcare/servlet/kundeservice_faq_linux) and it works fine.
Just to put down an additional comment - the procedure as described above for wvdial was not so troublefree, as I first thought. The problem is that when the modem is plugged in, there are in fact three usb serial ports created, which can be confirmed by: $ ls -la /dev/ttyU* crw-rw---- 1 root dialout 188, 2 2009-08-10 12:23 /dev/ttyUSB0 (first) crw-rw---- 1 root dialout 188, 3 2009-08-10 12:23 /dev/ttyUSB1 (second) crw-rw---- 1 root dialout 188, 4 2009-08-10 12:23 /dev/ttyUSB2 (third) Another problem is that if the USB modem was connected once, then disconnected, there are occasionally "hanging" USB serial ports, say /dev/ttyUSB0. Even if this "hanging" port is deleted using rm, upon second connection, the enumeration of the three ports will start from the next after the last "dangling" one - that is, even after deleting the "hanging" port, plugging in the USB modem afterwards for the second time will result with: $ ls -la /dev/ttyU* crw-rw---- 1 root dialout 188, 2 2009-08-10 12:23 /dev/ttyUSB1 (first) crw-rw---- 1 root dialout 188, 3 2009-08-10 12:23 /dev/ttyUSB2 (second) crw-rw---- 1 root dialout 188, 4 2009-08-10 12:23 /dev/ttyUSB3 (third) Correspondingly, changes need to be made to /etc/wvdial.conf. This is where the biggest problem was for me - I thought that the port that wvdial should connect to is the last one; that is, the (third) port in this case. Upon such a setting, wvdial would manage to connect to the network for some 30 secs, and then disconnect with errors: LCP: timeout sending Config-Requests, Receive serial link is not 8-bit clean: Problem: all had bit 7 set to 0 (these errors can be seen in /var/log/syslog [NOT /var/log/messages], once ppp debug is enabled (sudo nano /etc/ppp/options) ) Inspecting /var/log/syslog also pointed to the solution - that is, the USB serial port one should use in /etc/wvdial.conf is not the last, but instead the (first) one; as syslog will point out: Aug 10 14:35:30 mycomputer NetworkManager: <info> (ttyUSB3): ignoring due to lack of mobile broadband capabilties Aug 10 14:35:30 mycomputer NetworkManager: <info> (ttyUSB1): found serial port (udev:GSM hal:GSM) Aug 10 14:35:30 mycomputer NetworkManager: <info> (ttyUSB1): new Modem device (driver: 'option' Aug 10 14:35:30 mycomputer NetworkManager: <info> (ttyUSB2): ignoring due to lack of mobile broadband capabilties In which case, ttyUSB1 should be the USB serial port used in /etc/wvdial.conf. With this, the full procedure to get the E1550 running on Ubuntu 9.04, EEE PC 900 would be: - In Terminal: tail -f /var/log/syslog - Plug in the USB modem - Follow the messages in syslog, see which ttyUSB port is to be used for the modem - sudo nano /etc/wvdial.conf, and make sure the settings refer to the port in previous step, save and close - In Network Manager, the broadband connection should appear once the modem is detected; activate the connection and wait for it to fail - in Terminal: sudo wvdial So far, I haven't had problems since I started using this procedure - hope it stays that way :) Cheers!
you actually want to completely stop NetworkManager, then run this in a terminal, as root: NM_SERIAL_DEBUG=1 NM_PPP_DEBUG=1 /usr/sbin/NetworkManager --no-daemon the 'restart' argument you mentioned above isn't actually an NM option and wont' give the right debugging information. Try to connect, then paste the output from NM into this bug report. Thanks!
Hi Dan, Thanks for your response.. I was about to make a generate debug info for this bug, when suddenly I realized, that now this modem connects correctly on at least one of my Ubuntu machines!! Which is strange, as I cannot remember Network Manager being updated recently (but I could be wrong :) ) In any case, seems it works fine now - if I ever get a problem again with it, I'll report. Thanks !! Cheers...
Created attachment 149477 [details] E1552_NetworkManger.log
Created attachment 149478 [details] E1552_daemon.log
Created attachment 149479 [details] E1552_HOME_dot_wvdial.conf (gnome-ppp)
Hi, i also got an E1552 modem, which is probably similar to the one described in this bug. I'm using Ubuntu-9.10. I bought this one, because my former ZTE MF628 doesn't work at all. lsusb: ID 12d1:1446 Huawei Technologies Co., Ltd. lsusb (after about 10 seconds) ID 12d1:1001 Huawei Technologies Co., Ltd. E620 USB Modem The problems i got is exactly the same as described here. The modem regularly fails to connect, but rarely it succeeds. I believe the important message is "Got failure code 3: No carrier" in the daemon.log (from modem-manager). With gnome-ppp the device works, and i was also able to switch the modem to "always 3G" mode, which was not possible with NetworkNanager (the line Init3 = AT^SYSCFG=14,2,3FFFFFFF,2,4 did that). Before, it always grabbed GPRS connections, which - of course - sucked speed-wise. Attachments: 1) output of NetworkManager 2) daemon.log 3) $HOME./.wvdial.conf. (used by gnome-ppp) The logs show a number of failed attempts and one successful one... Btw, it would be really nice if NetworkManager would show a bit more info about the available networks (signal-strength, GPRS vs UMTS), like "umtsmon" does.
Ok - i found out how to enable debug logging in modem-manager (sudo kill -SIGUSR1 <PID>) the relevant output: Dec 9 22:14:25 one modem-manager: (ttyUSB0) opening serial device... Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'ATZ E0 V1 X4 &C1 +CMEE=1<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'ATE0 +CMEE=1<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CREG=0<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CFUN=1<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CPIN?<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>+CPIN: READY<CR><LF><CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+COPS=0,,<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CREG?<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>+CREG: 0,5<CR><LF><CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CGDCONT?<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>+CGDCONT: 1,"IP","web.yesss.at","0.0.0.0",0,0<CR><LF>+CGDCONT: 2,"IP","web.yesss.at","0.0.0.0",0,0<CR><LF><CR><LF>OK<CR><LF>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'ATD*99***1#<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>NO CARRIER<CR><LF>' Dec 9 22:14:25 one modem-manager: Got failure code 3: No carrier Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CEER<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>+CEER: No cause information available<CR><LF><CR><LF>OK<CR><LF>' then i took the wvdial.conf from gnome-ppp and added the AT+... commands from modem-manger one by one. (wvdial -C mywvdial.conf) Where it stopped working was the line: "AT+COPS=0,," Regards, Norbert
The +COPS=0,, appears fine because the modem returned OK. However, the registration status request in the next line: Dec 9 22:14:25 one modem-manager: (ttyUSB0): --> 'AT+CREG?<CR>' Dec 9 22:14:25 one modem-manager: (ttyUSB0): <-- '<CR><LF>+CREG: 0,5<CR><LF><CR><LF>OK<CR><LF>' indicates that the network the modem is registered on a roaming network. And the NO CARRIER indicates that the data call was not able to be set up. Perhaps the roaming network doesn't actually accept your APN? Ubuntu 9.10 ships a rather old snapshot of ModemManager too, and we can get more information out of the device with newer versions of MM (including getting the actual network name). You might want to check out the NetworkManager PPA here: https://edge.launchpad.net/~network-manager/+archive/trunk which contains nightly builds of the latest MM and NM from git.
Hi Dan, I upgraded the packages, but unfortunately no real change in behaviour. Dec 11 02:11:17 daisy modem-manager: (ttyUSB0) opening serial device... Dec 11 02:11:17 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/0: state changed (disabled -> enabling) Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): --> 'ATZ E0 V1 +CMEE=1<CR>' Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): --> 'ATE0 +CMEE=1<CR>' Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): --> 'ATX4 &C1<CR>' Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:11:17 daisy modem-manager: (ttyUSB0): --> 'AT+CREG=0<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'AT+CFUN=1<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/0: state changed (enabling -> registered) Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'AT+CPIN?<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CPIN: READY<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'AT+COPS=0,,<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'AT+CREG?<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CREG: 0,5<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'AT+CGDCONT?<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CGDCONT: 1,"IP","web.yesss.at","0.0.0.0",0,0<CR><LF>+CGDCONT: 2,"IP","web.yesss.at","0.0.0.0",0,0<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/0: state changed (registered -> connecting) Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'ATD*99***1#<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>NO CARRIER<CR><LF>' Dec 11 02:11:18 daisy modem-manager: Got failure code 3: No carrier Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): --> 'AT+CEER<CR>' Dec 11 02:11:18 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CEER: No cause information available<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:11:18 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/0: state changed (connecting -> registered) It's really weird, then I tried sending AT+COPS? with a custom wvdial.conf and got the response: +COPS: 1,2,"23205",2 When I enter this "23205" into the "Network" field of the Network Connections/Mobile Broadband, the modem works. A successful communication looks like this: Dec 11 02:57:04 daisy modem-manager: (ttyUSB0) opening serial device... Dec 11 02:57:04 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/1: state changed (disabled -> enabling) Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'ATZ E0 V1 +CMEE=1<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'ATE0 +CMEE=1<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'ATX4 &C1<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'AT+CREG=0<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'AT+CFUN=1<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/1: state changed (enabling -> registered) Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'AT+CPIN?<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CPIN: READY<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'AT+COPS=1,2,"23205"<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'AT+CREG?<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CREG: 0,5<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'AT+CGDCONT?<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>+CGDCONT: 1,"IP","web.yesss.at","0.0.0.0",0,0<CR><LF>+CGDCONT: 2,"IP","web.yesss.at","0.0.0.0",0,0<CR><LF><CR><LF>OK<CR><LF>' Dec 11 02:57:05 daisy modem-manager: Modem /org/freedesktop/ModemManager/Modems/1: state changed (registered -> connecting) Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): --> 'ATD*99***1#<CR>' Dec 11 02:57:05 daisy modem-manager: (ttyUSB0): <-- '<CR><LF>CONNECT 3600000<CR><LF>'
Created attachment 149560 [details] Complete log (starting from plugging in the modem) Perhaps this part is interesting: --> AT+COPS=3,2;+COPS? <-- +COPS: 0,2,"23205",2 --> AT+COPS=3,0;+COPS? <-- +COPS: 0,0,"one",2 (yes, it's probably a roaming account i got - "Yesss" is a subcontractor of "one".
I wonder if sending "+COPS=0,," is a good idea when the user hasn't entered a "Network" parameter. This patch fixes the problem for me, and the modem will always connect - even without setting "Network". diff --git a/src/mm-generic-gsm.c b/src/mm-generic-gsm.c index 55d75f6..149b45d 100644 --- a/src/mm-generic-gsm.c +++ b/src/mm-generic-gsm.c @@ -1203,7 +1244,10 @@ do_register (MMModemGsmNetwork *modem, if (network_id) command = g_strdup_printf ("+COPS=1,2,\"%s\"", network_id); else - command = g_strdup ("+COPS=0,,"); + { + /* command = g_strdup ("+COPS=0,,"); */ + command = g_strdup ("AT"); /* send dummy command, "+COPS=0,," seems to cause "no carrier" problems */ + } /* Ref the callback info to ensure it stays alive for register_done() even * if the timeout triggers and ends registration (which calls the callback
Interesting; unfortunately +COPS=0,, is what's supposed to make the modem ensure it's registered on the default/home network... thanks for looking into that, I'll do some more testing with my E160G and see if I can come up with something. GSM-07.07: +COPS=[<mode>[,<format>[,<oper>]]] "Set command forces an attempt to select and register the GSM network operator. <mode>: 0 automatic (<oper> field is ignored) " So +COPS=0,, directs the modem to register automatically on the network it thinks is best. If you want to direct the modem to register only on a specific network you can enter that MCC/MNC into the connection editor's "Network" box and then the code will do +COPS=1,2,"310260" for example and force registration to only that network.
(In reply to comment #13) > Interesting; unfortunately +COPS=0,, is what's supposed to make the modem > ensure it's registered on the default/home network... Probably it's not really specified how the modem is supposed to decide in "automatic-mode". The modem could also toss a coin or decide upon the signal strength. ;-) Therefore I guess three "Network" options in the Wireless Config would be cool: 1) Never change a winning team. (does nothing or sends a dummy "AT") 2) Let the modem decide. Sends "AT+COPS=0,," 3) Let the user decide (Text-Box or better drop-down of available operators). Sends "AT+COPS=1,2,..."
Created attachment 155903 [details] mm log for E1552 (git master) connection attempt failed with "no carrier"
Created attachment 155905 [details] mm log for E1552 (git master) connecting works (without sending AT+COPS=0,,) I tried it again. With sending a dummy "AT" rather than "AT+COPS=0,," the problem goes away... (see Comment #12) Maybe a bug in the E1552 firmware?
Created attachment 156057 [details] As Comparison: Same SIM as in Comments 15/16, but with the ZTE MF626 (works) The difference between the two Modems seems to be that the Huawai E1552 is set to roaming mode (Is this a modem setting?) Huawai E1552: ** (modem-manager:7958): DEBUG: (ttyUSB0): --> 'AT^SYSCFG?<CR>' ** (modem-manager:7958): DEBUG: (ttyUSB0): <-- '<CR><LF>^SYSCFG:2,2,3FFFFFFF,1,2<CR><LF><CR><LF>OK<CR><LF>' The fourth field apparently describes the roaming setting (0=not supported | 1=roaming) ** (modem-manager:7958): DEBUG: (ttyUSB0): --> 'AT+CREG?<CR>' ** (modem-manager:7958): DEBUG: (ttyUSB0): <-- '<CR><LF>+CREG: 2,5,0A,9A22<CR><LF><CR><LF>OK<CR><LF>' ** (modem-manager:7958): DEBUG: (ttyUSB0): --> 'AT+COPS=3,2;+COPS?<CR>' ** (modem-manager:7958): DEBUG: (ttyUSB0): <-- '<CR><LF>+COPS: 0,2,"23205",2<CR><LF><CR><LF>OK<CR><LF>' ** (modem-manager:7958): DEBUG: (ttyUSB0): --> 'AT+COPS=3,0;+COPS?<CR>' ** (modem-manager:7958): DEBUG: (ttyUSB0): <-- '<CR><LF>+COPS: 0,0,"one",2<CR><LF><CR><LF>OK<CR><LF>' ** (modem-manager:7958): DEBUG: (ttyUSB0): --> 'AT+CSQ<CR>' ZTE MF626: ** (modem-manager:4238): DEBUG: (ttyUSB2): --> 'AT+CREG?<CR>' ** (modem-manager:4238): DEBUG: (ttyUSB2): <-- '<CR><LF>+CREG: 1,1<CR><LF><CR><LF>OK<CR><LF>' ** (modem-manager:4238): DEBUG: (ttyUSB2): --> 'AT+COPS=3,2;+COPS?<CR>' ** (modem-manager:4238): DEBUG: (ttyUSB2): <-- '<CR><LF>+COPS: 0,2,"23205",2<CR><LF><CR><LF>OK<CR><LF>' ** (modem-manager:4238): DEBUG: (ttyUSB2): --> 'AT+COPS=3,0;+COPS?<CR>' ** (modem-manager:4238): DEBUG: (ttyUSB2): <-- '<CR><LF>+COPS: 0,0,"Orange Austria",2<CR><LF><CR><LF>OK<CR><LF>' ** (modem-manager:4238): DEBUG: (ttyUSB2): --> 'AT+CSQ<CR>' --------- +CREG: 0,0 - The GT is not registered or scanning for a GSM network +CREG: 0,1 - The GT has registered on the "HOME" network of the SIM +CREG: 0,2 - The GT is not registered but is scanning for a GSM network +CREG: 0,3 - Registration is denied (Manual registration attempt failed) +CREG: 0,5 - The GT has registered on to another network (it is roaming). --------- Perhaps the result of +COPS=0,, is undefined in roaming mode... ... and whats the point of sending +COPS=0,, anyway, if AT+COPS? already returns an operator (and the user hasn't manually specified a different one).
Hmm .. turning off roaming with minicom didn't help either... AT^SYSCFG=2,2,3FFFFFFF,0,2 It still reports "roaming" mode. I also tried AT^SYSCFG=14,2,3FFFFFFF,0,2 but no improvement ^SYSCFG=mode, order, band, roaming, domain Mode 2 Automatic search 13 2G ONLY 14 3G ONLY 16 No change Order 0 Automatic search 1 2G first, then 3G 2 3G first, then 2G 3 No change Band 80 GSM DCS systems 100 Extended GSM 900 200 Primary GSM 900 200000 GSM PCS 400000 WCDMA IMT 2000 3FFFFFFF Any band 40000000 No change of band Roaming 0 Not supported 1 Roaming is supported 2 No change Domain 0 CS_ONLY 1 PS_ONLY 2 CS_PS 3 ANY 4 No change
just for the record (my E1552) AT+GMR 11.608.12.00.00 OK AT+GMI huawei OK
*** Bug 613897 has been marked as a duplicate of this bug. ***
Any chance I can get an AT+GMM out of the modem? That's what we'd be using for the special-casing, if any. In any case, the point about not sending the COPS=0,, when the modem is already registered is a good one.
For the don't-send-COPS=0-when-registered thing, see: 4d426ebbbd210795e31869225f895257c3546112 That won't help much if you try to make a connection right after plugging the modem in (before it's had time to complete it's auto registration) but it'll help if you wait a bit before connecting. Once I get +GMM results for the E1552 we can special-case it perhaps. Though I'm a bit concerned about the dup-ed report from the 1550.
As the guy with the E1550, is there anything I can do to contribute? While I have some admin experience, I never had anything to do with modems. I suppose AT+GMM means sending the Huawei a command and report the answer? Is 'minicom' the program to do this? (A mini-tutorial what to type would be most welcome) For the record, I had seen 591047 before filing 613897, but 591047 was for version 0.7 whereas I noticed the problem with 0.8. Since the two versions even have different debug instructions, I ended up filing a new bug report.
(In reply to comment #22) > For the don't-send-COPS=0-when-registered thing, see: > > 4d426ebbbd210795e31869225f895257c3546112 > > That won't help much if you try to make a connection right after plugging the > modem in (before it's had time to complete it's auto registration) but it'll > help if you wait a bit before connecting. > Yeah! I have just tested the current git master with my E1552 and it works well now :-) (I also tested with the ZTE MF626 and MF628, they work well too - plus connecting seems to be faster now). Btw, i had a look at the E1552 modem-software log-files on WinXP: It doesn't send AT+COPS=0,, either. What's weird is that this modem is always operating in roaming mode. Even on Windows. The ZTE modems work in "normal" non-roaming mode with the same SIM cards. So perhaps always switching to roaming is the actual bug in the firmware (but perhaps an opportunity to test roaming behaviour, which would be difficult otherwise). Perhaps the problem could also be resolved differently - with a dialog which pops up on connection failure (no carrier) - prompting to select from the (scanned) list of available networks (because specifying the network solves the problem too). Such a dialog is a missing feature in MM/NM anyway.
Info for the E1550 case; will look into compiling network manager from source soon, given the above encouraging report. AT+GMR 11.608.13.00.47 AT+GMI huawei AT+GMM E1550 Hope that helps ...
Eh. I'm reopening although I'm not too sure the requested information has been provided. Feel free to close right away.
Reopen if still applicable.