Probably has to wait for ModemManager since it's highly device dependent.  Other cards I've seen handle the errors differently, or the error simply doesn't refer to the SIM at all.

(In reply to comment #1)
> Probably has to wait for ModemManager since it's highly device dependent. 
> Other cards I've seen handle the errors differently, or the error simply
> doesn't refer to the SIM at all.

Still not fixed with ModemManager:

Latest modemmanager master branch gives in log:

Feb 2 20:21:24 marius-T1005 ModemManager[1105]: <warn> Modem couldn't be initialized: Couldn't check unlock status: SIM failure

but there is no indication to the user (nm-applet behaves as if mobile broadband is unavailable).

Seems I get (from "src/mm-error-helpers.c"):

    { MM_MOBILE_EQUIPMENT_ERROR_SIM_FAILURE, "simfailure", "SIM failure" },

but should have got:

{ MM_MOBILE_EQUIPMENT_ERROR_SIM_NOT_INSERTED, "simnotinserted", "SIM not inserted" }

Anyway (any of them should be enough hint to the user), it seems like the error state is not propagated (src/mm-broadband-modem.c):

static void
iface_modem_initialize_ready (MMBroadbandModem *self,
                              GAsyncResult *result,
                              InitializeContext *ctx)
{
    GError *error = NULL;

    /* If the modem interface fails to get initialized, we will move the modem
     * to a FAILED state. Note that in this case we still export the interface. */
    if (!mm_iface_modem_initialize_finish (MM_IFACE_MODEM (self), result, &error)) {
        /* Report the new FAILED state */
        mm_warn ("Modem couldn't be initialized: %s", error->message);
        g_error_free (error);

        mm_iface_modem_update_state (MM_IFACE_MODEM (self),
                                     MM_MODEM_STATE_FAILED,
                                     MM_MODEM_STATE_CHANGE_REASON_UNKNOWN);

        /* Jump to the firmware step. We allow firmware switching even in failed
         * state */
        ctx->step = INITIALIZE_STEP_IFACE_FIRMWARE;
        initialize_step (ctx);
        return;
    }

Downstream bug report: https://bugs.launchpad.net/network-manager/+bug/1113703

I tested using modem manager 0.6 (Ubuntu Precise) now with an external USB modem, where the missing SIM is not detected at init and thus the device is in the modem device list and it is possible to try connecting. When I do:

MM reports:

NetworkManager[5163]: <info> Activation (ttyUSB0) Stage 1 of 5 (Device Prepare) scheduled...
NetworkManager[5163]: <info> Activation (ttyUSB0) Stage 1 of 5 (Device Prepare) started...
NetworkManager[5163]: <info> Activation (ttyUSB0) Stage 1 of 5 (Device Prepare) complete.
NetworkManager[5163]: <warn> GSM modem enable failed: (32) Failed to find a usable modem character set
NetworkManager[5163]: <debug> [1360004871.251768] [nm-modem-gsm.c:150] translate_mm_error(): unmapped dbus error detected: 'org.freedesktop.ModemManager.Modem.UnsupportedCharset'
NetworkManager[5163]: <info> (ttyUSB0): device state change: prepare -> failed (reason 'modem-init-failed') [40 120 28]
NetworkManager[5163]: <debug> [1360004871.252465] [nm-manager.c:512] manager_device_state_changed(): stopping connectivity checks
NetworkManager[5163]: <warn> Activation (ttyUSB0) failed for connection 'Netcom Default'
NetworkManager[5163]: <debug> [1360004871.253379] [nm-device.c:4626] nm_device_queue_state(): (ttyUSB0): queued state change to disconnected (id 68)
NetworkManager[5163]: <debug> [1360004871.254205] [nm-device.c:4594] queued_set_state(): (ttyUSB0): running queued state change to disconnected (id 68)

And Network manager reports:

modem-manager[3823]: <debug> [1360004871.202712] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): --> 'AT^SYSCFG?<CR>'
modem-manager[3823]: <debug> [1360004871.217798] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): <-- '<CR><LF>^SYSCFG:2,0,3FFFFFFF,1,2<CR><LF><CR><LF>OK<CR><LF>'
modem-manager[3823]: <debug> [1360004871.218155] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): --> 'AT+CLCK=?<CR>'
modem-manager[3823]: <debug> [1360004871.231332] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): <-- '<CR><LF>+CME ERROR: 13<CR><LF>'
modem-manager[3823]: <debug> [1360004871.231572] [mm-serial-parsers.c:448] mm_serial_parser_v1_parse(): Got failure code 13: SIM failure
modem-manager[3823]: <debug> [1360004871.231668] [mm-serial-port.c:908] mm_serial_port_close(): (ttyUSB2) device open count is 1 (close)
modem-manager[3823]: <debug> [1360004871.231828] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): --> 'AT+CSCS=?<CR>'
modem-manager[3823]: <debug> [1360004871.244115] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): <-- '<CR><LF>+CME ERROR: 13<CR><LF>'
modem-manager[3823]: <debug> [1360004871.244339] [mm-serial-parsers.c:448] mm_serial_parser_v1_parse(): Got failure code 13: SIM failure
modem-manager[3823]: <debug> [1360004871.244566] [mm-at-serial-port.c:334] debug_log(): (ttyUSB2): --> 'AT+CMER=3,0,0,1<CR>'
modem-manager[3823]: <info>  [1360004871.245531] [mm-modem.c:764] mm_modem_set_state(): Modem /org/freedesktop/ModemManager/Modems/1: state changed (enabling -> disabled)
modem-manager[3823]: <debug> [1360004871.245635] [mm-serial-port.c:908] mm_serial_port_close(): (ttyUSB2) device open count is 0 (close)
modem-manager[3823]: <info>  [1360004871.245700] [mm-serial-port.c:924] mm_serial_port_close(): (ttyUSB2) closing serial port...
modem-manager[3823]: <info>  [1360004871.246878] [mm-serial-port.c:956] mm_serial_port_close(): (ttyUSB2) serial port closed

Thus it seems like the information about the SIM failure is lost by NM, or shadowed by the first unimportant error?

Then nm-applet looses even more information by just saying connection error.

Seems like MM sends a symptom error message instead of the returned error message. In src/mm-generic-gsm.c:enabled_set_charset_done(...):

       if (best_charsets[idx] == MM_MODEM_CHARSET_UNKNOWN) {
            GError *tmp_error;

            /* No more character sets we can use */
            tmp_error = g_error_new_literal (MM_MODEM_ERROR,
                                             MM_MODEM_ERROR_UNSUPPORTED_CHARSET,
                                             "Failed to find a usable modem character set");
            enable_failed (modem, tmp_error, info);
            g_error_free (tmp_error);

I do not understand why it doesn't stop at the response "+CME ERROR: 13" to "AT+CSCS=?". Anyway, I think MM should return all unexpected errors directly to NM.

Now that bug #691778 is solved, is there any more to do in network manager or is it just updating of the GUI clients left?

(In reply to comment #5)
> Now that bug #691778 is solved, is there any more to do in network manager or
> is it just updating of the GUI clients left?

Well, NM needs to avoid skipping modems in Failed state, which is something that we currently do when using MM >= 0.7. Have a branch for that already, just need to test it.

bugzilla.gnome.org is being shut down in favor of a GitLab instance. 
We are closing all old bug reports and feature requests in GNOME Bugzilla which have not seen updates for a long time.

If you still use NetworkManager and if you still see this bug / want this feature in a recent and supported version of NetworkManager, then please feel free to report it at https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/

Thank you for creating this report and we are sorry it could not be implemented (workforce and time is unfortunately limited).