Re: Changing Auto-Connect feature
On Thu, 2010-05-27 at 05:56 +0200, Frederik Nnaji wrote: > On Tue, May 25, 2010 at 18:45, Franco Miceli > wrote: > Therefore we want to to make some changes to the NM code in > order to take into consideration other factors such as: SNR, > loss %, etc. > > > will you considered working on a UI for manually selecting and > configuring a symbolic visual representation of known APs currently > within range/sight? > This would also help with GeoLocation, since it could use "strength" > to approximate My position by the nearness of a known AP currently > within sight. We've thought about strength for a while, but in the end it doesn't work very well. There are simply too many things that can affect AP signal strength to use it as a reliable measure of where you are. It's better to use *more than one* AP and triangulate to get better accuracy than just 100m radius of one AP's known location. Using only one AP, if somebody turns on a microwave or uses a 2.4GHz DECT phone, suddenly it looks like you're 30 meters farther away from the AP than you were 10 seconds ago... Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: IP4Config and routes
On Wed, 2010-06-02 at 09:51 +0200, Simon Schampijer wrote: > On 12/18/2009 03:14 PM, Daniel Drake wrote: > > On Thu, 2009-12-17 at 14:22 -0800, Dan Williams wrote: > >> What would you expect the routing table to look like in your case? I > >> suppose we could do a default route for link-local. Not sure if that > >> will confuse apps that expect a default route to mean an internet > >> connection though. > > > > I would expect the subnet route, as NM is creating already: > > dest=169.254.0.0 > > gateway=0.0.0.0 > > genmask=255.255.0.0 > > > > I would also like the routing table to either include a default route: > > dest=0.0.0 > > gateway=0.0.0.0 > > genmask=0.0.0.0 > > > > or a multicast one: > > dest=224.0.0.0 > > gateway=0.0.0.0 > > genmask=240.0.0.0 > > > > The routing table that NM is setting up now is reasonable, in my > > opinion, but there should be some way of customizing the behaviour in > > the settings object. > > > > Daniel > > Hi, > > what is the status on this one? Was there a conclusion on whether NM > should set a default route for link local? Creating a multicast route by default on link-local IPv4 connections seems reasonable. Want to do a patch for that? I'd say just add the desired route in aipd_get_ip4_config() in src/nm-device.c to the returned NMIP4Config object and then lets make sure the code that adds routes works correctly there. Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: access to "ppp0" on dbus
On Wed, 2010-06-02 at 09:48 +0200, Markus Becker wrote: > Hi, > > is it somehow possible to get the ppp device of a mobile broadband connection > (e.g. ppp0), which was created by a dial-in connection, over dbus? > > Up to now we only found the physical device name (ttyUSB0) and the Udi. Not yet, but I'll probably apply a patch soon that will expose that as the "IpInterface" property of a device, which should contain what you need when the device reached the ACTIVATED state (as determined by it's State property). Dan > Thanks, > Markus > > > | ATTENTION: NEW TELEPHONE EXTENSION! > > | Dipl.-Ing. Markus Becker > | Communication Networks > | Mobile Research Center > | TZI - Center for Computing Technologies > | University Bremen > | Germany > > ___ > networkmanager-list mailing list > networkmanager-list@gnome.org > http://mail.gnome.org/mailman/listinfo/networkmanager-list ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: UI for dealing with certs appears insecure
On Fri, 2010-06-04 at 10:49 -0700, L. David Baron wrote: > I regularly want to access a particular wireless network using a > password whose security I care about, but I currently avoid using > this network because of security concerns I describe here. This > wireless network has what appears to me to be a reasonable > authentication mechanism for the desired level of security: it's > using WPA2 Enterprise with EAP-MSCHAPv2 authentication (I hope I'm > using the right terminology here), and has a valid cert signed by a > well-known root CA, for a hostname that makes sense in context > (i.e., its domain is the domain of the company operating the > wireless network). > > However, the NetworkManager UI doesn't give me confidence in the > handling of the security of my password, since it prompts me for > (all at once): > * my username > * my password > * what root CA should be used (if any) to validate the cert > > In this particular case, it seems somebody could steal my password > if they set up a wireless network nearby with the same SSID, a > stronger signal, and a valid cert purchased from the same CA (but > for a different domain). Or, if I choose the full root cert list > for the CA (since I really don't know any other way to figure out > what the right root CA is other than finding a friend with a Mac to > connect to that wireless network), the attacker could use a valid > cert from any CA. > > It seems to me that in cases where certificates are involved: > * the prompt for my username and password should not happen until >the cert has been checked, and it should display information >about the cert, i.e.: >+ if the cert was signed by a CA in the root cert list, the > hostname the cert is for (and probably the CA that signed it, > and perhaps the fingerprint) >+ otherwise, the cert's fingerprint > * when I enter a username and password for such a prompt, it should >only be used for wireless with that SSID+Cert combination, and >not for other wireless networks with the same SSID. > > It seems like this would prevent the attack described above, and > also improve security in the self-signed cert case. A UI that > worked this way would make me comfortable accessing the network in > question using NetworkManager. > > > Is this a reasonable request? Are there reasons the current UI is > preferable to such a UI? This is a reasonable request and making the certificate UI better is something that I've wanted to do for a long time. Just a month or two or three ago Jouni added better cert validation results to wpa_supplicant, which means we now actually have a chance of getting better status about 802.1x authentication. Before, we'd have had to run wpa_supplicant in verbose mode and screenscraped its output (which was really OpenSSL error messages) to find out what was going on. I'd like to use this eventually (though wpa_supplicant isn't there yet) to prompt the user to accept the RADIUS sever's certificate if it's not been seen yet like Mac OS X and Windows do, and then save the fingerprint like you suggest and warn the user if that fingerprint changes. We're almost there, but we need a bit more intelligence underneath to do so. Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: RS232 GSM Modem
On Wed, 2010-06-02 at 12:09 +0200, Tom wrote: > On Wed, 2010-06-02 at 00:54 -0700, Dan Williams wrote: > > Hmm, what are you using to connect with PPP? If you're using > > NetworkManager, you can: > > > > NM_PPP_DEBUG=1 NetworkManager --no-daemon > > > > and get verbose PPP output which could help to debug the issue. > > > > Note that PPP uses it's *own* baudrate, which you have to send to pppd > > on the command line. So if we dont' set that correctly, PPP may not > > work. But debug output from pppd including it's command-line options > > might help us figure that out. > > i set the baudrate for ppp (in my system-settings connection). The > output of NetworkManager (0.8 from debian testing repository) follows. > In this case i used the modem with a usb<->adapter on my Lenovo R400 > laptop. But this should work, too. Looks like the remote side doesn't like use of Compression Control Protocol. Try: nobsdcomp=true nodeflate=true no-vj-comp=true And see if that works? Dan > > > NetworkManager: Activation (ttyUSB1) starting connection 'TC63' > NetworkManager: (ttyUSB1): device state change: 3 -> 4 (reason > 0) > NetworkManager: Activation (ttyUSB1) Stage 1 of 5 (Device > Prepare) scheduled... > NetworkManager: Activation (ttyUSB1) Stage 1 of 5 (Device > Prepare) started... > NetworkManager: Activation (ttyUSB1) Stage 1 of 5 (Device > Prepare) complete. > NetworkManager: Activation (ttyUSB1) Stage 2 of 5 (Device > Configure) scheduled... > NetworkManager: Activation (ttyUSB1) Stage 2 of 5 (Device > Configure) starting... > NetworkManager: (ttyUSB1): device state change: 4 -> 5 (reason > 0) > NetworkManager: Activation (ttyUSB1) Stage 2 of 5 (Device > Configure) successful. > NetworkManager: Activation (ttyUSB1) Stage 3 of 5 (IP Configure > Start) scheduled. > NetworkManager: Activation (ttyUSB1) Stage 2 of 5 (Device > Configure) complete. > NetworkManager: Activation (ttyUSB1) Stage 3 of 5 (IP Configure > Start) started... > NetworkManager: (ttyUSB1): device state change: 5 -> 7 (reason > 0) > NetworkManager: Starting pppd connection > NetworkManager: [1275473061.909184] nm_ppp_manager_start(): > Command line: /usr/sbin/pppd nodetach lock nodefaultroute debug user > vodafone ttyUSB1 noipdefault 57600 noauth usepeerdns lcp-echo-failure 0 > lcp-echo-interval 0 ipparam /org/freedesktop/NetworkManager/PPP/0 > plugin /usr/lib/pppd/2.4.4/nm-pppd-plugin.so > NetworkManager: [1275473061.971552] nm_ppp_manager_start(): ppp > started with pid 7242 > NetworkManager: Activation (ttyUSB1) Stage 4 of 5 (IP6 Configure > Get) scheduled... > NetworkManager: Activation (ttyUSB1) Stage 3 of 5 (IP Configure > Start) complete. > NetworkManager: Activation (ttyUSB1) Stage 4 of 5 (IP6 Configure > Get) started... > NetworkManager: Activation (ttyUSB1) Stage 4 of 5 (IP6 Configure > Get) complete. > Plugin /usr/lib/pppd/2.4.4/nm-pppd-plugin.so loaded. > ** Message: nm-ppp-plugin: (plugin_init): initializing > ** Message: nm-ppp-plugin: (nm_phasechange): status 3 / phase 'serial > connection' > using channel 1 > Using interface ppp0 > Connect: ppp0 <--> /dev/ttyUSB1 > ** Message: nm-ppp-plugin: (nm_phasechange): status 5 / phase > 'establish' > sent [LCP ConfReq id=0x1 > ] > NetworkManager:SCPlugin-Ifupdown: devices added > (path: /sys/devices/virtual/net/ppp0, iface: ppp0) > NetworkManager:SCPlugin-Ifupdown: device added > (path: /sys/devices/virtual/net/ppp0, iface: ppp0): no ifupdown > configuration found. > rcvd [LCP ConfNak id=0x1 ] > sent [LCP ConfReq id=0x2 > ] > rcvd [LCP ConfAck id=0x2 > ] > rcvd [LCP ConfReq id=0x3 > ] > sent [LCP ConfAck id=0x3 > ] > ** Message: nm-ppp-plugin: (nm_phasechange): status 6 / phase > 'authenticate' > ** Message: nm-ppp-plugin: (get_credentials): passwd-hook, requesting > credentials... > ** Message: nm-ppp-plugin: (get_credentials): got credentials from > NetworkManager > sent [PAP AuthReq id=0x1 user="vodafone" password=] > rcvd [PAP AuthAck id=0x1 "TTP Com PPP - Password Verified OK"] > Remote message: TTP Com PPP - Password Verified OK > PAP authentication succeeded > ** Message: nm-ppp-plugin: (nm_phasechange): status 8 / phase 'network' > sent [CCP ConfReq id=0x1 ] > sent [IPCP ConfReq id=0x10.0.0.0> ] > rcvd [LCP ProtRej id=0x0 80 fd 01 01 00 0f 1a 04 78 00 18 04 78 00 15 03 > 2f 32] > Protocol-Reject for 'Compression Control Protocol' (0x80fd) received > rcvd [LCP TermReq id=0x0 "Normal Termination by NCP"] > LCP terminated by peer (Normal Termination by NCP) > ** Message: nm-ppp-plugin: (nm_phasechange): status 5 / phase > 'establish' > sent [LCP TermAck id=0x0] > ** Message: nm-ppp-plugin: (nm_phasechange): status 11 / phase > 'disconnect' > Connection terminated. > NetworkManager: (ttyUSB1): device state change: 7 -> 9 (reason > 13) > NetworkManager: Marking connection 'TC63' invalid. > NetworkManager: Activation (ttyUSB1) failed. > NetworkManager: (ttyUSB1): device state change: 9 -> 3 (reas
Re: [RFC] Fast-user-switching plans
On Wed, 2010-06-02 at 23:44 +1200, Simon Geard wrote: > On Wed, 2010-06-02 at 00:45 -0700, Dan Williams wrote: > > The one benefit of user connections is that they follow you if you back > > up your homedir and switch machines :) I don't think that's enough of a > > benefit to keep them around though. > > Also if you periodically update your OS - e.g installing a new Ubuntu > release every six months. Stuff in $HOME stays - stuff in /etc doesn't. Yup. Though we could mitigate that by providing usable backup/restore capability that dumps a bunch of keyfiles describing your connections into a tarball or something. Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: dhcpcd timeout
On Thu, 2010-06-03 at 10:34 +0400, Victor Gaydov wrote: > Hello, > > After upgrade from 0.7 to 0.8 NetworkManager compiled with dhcpcd can't > start connection because of dhcpcd timeout. Is it a way to change this > timeout? > > Notes: > - NetworkManager compiled with dhclient works fine > - dhcpcd executed manually work fine > - it seems that lease-time from /etc/dhcp/dhcpd.conf doesn't matter >when dhcpcd is executed by NM > > Log attached. Is there any chance you could run wireshark on the interface and see if any DHCP packets get out? This is quite odd and it would point to either a misconfiguraton of dhcpcd on NM's part (maybe sendign the wrong command-line arguments) or it could point to a problem in dhcpcd. It might also be useful to strace dhcpcd to see what it's doing. The logs show the problem, but don't really provide any insight into what might be going on. The other alternative may be to recompile NM and send any debugging or verbose command-line arguments that dhcpcd might use to get more output from it about the issue. Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: Programmatically set the MTU for ALL Ethernet Adapters
On Wed, 2010-06-02 at 11:37 -0400, Jeffersen Sylvia wrote: > I am running into an issue with programmatically changing the MTU > settings for an Ethernet adapter within SUSE Enterprise 11. > > > Issue: We have configured an OS to run within VMware Player that will > be transferred between multiple PCs on removable media. By nature (as > well as design) VMware will assign a new MAC Address (and therefore a > new virtual adapter) at each new PC, or location on the same PC. > Because we are using Lotus Notes 8.5.1, and Cisco AnnyConnect VPN > Tunneling, we need to set the MTU for the adapter *WITHOUT* user > intervention. > > > > When Using *ifup* we can change global settings in the > “/etc/sysconfig/network/config” and adapter based settings in > “/etc/sysconfig/network/ifcfg.template” files, however once the > adapter is created, it must be manually activated and set to DHCP > which requires “root” permissions. > > > > When using NetworkManager, the process is automated for the creation > and default activation of the eth adapter, however the MTU must be > set manually, per adapter by the user. > > > > I am trying to find a way (similar to adjusting the configurations of > config/ifcfg.template) that will programmatically set the MTU for ALL > new eth adapters to 1492. Can setting the MTU happen *after* the connection has been brought up? If so, you could use a dispatcher script (see 'man NetworkManager') and if the interface name (which is passed as an argument to that script) matches ethX then set the mtu to 1492 using ifconfig. But NM also has built-in MTU support. There are two cases here. First, NM creates "default" connections for any Ethernet device on the system if that device does not have configuration already, where that configuration is determined by which system settings plugins you have specified in /etc/NetworkManager/nm-system-settings.conf. If you have 'ifcfg-suse' there, NM tries to read in legacy ifcfg files. If you have 'keyfile' there, you can use configuration files in /etc/NetworkManager/system-connections. In the ifcfg case, putting the line MTU=1492 into an ifcfg file will work, while for keyfile 'mtu=1492' in the [802-3-ethernet] section will work. In both cases, do not specify a MAC address, and thus NM will apply the configuration to any ethernet adapter that happens to appear. Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: How to conditionally compile plug-in source file to only support dhcpcd or dhclient?
On Fri, 2010-06-04 at 21:50 +0800, Mu Qiao wrote: > Hi Dan, > > Gentoo NM plugin has to read dhcp hostname and clientid from > configuration file of dhcpcd or dhclient. > > Currently I just test whether the configuration file of dhcpcd or > dhclient exists to determine which one is used. Could I make use of the > switch "--with-dhcpcd=yes" to enable conditional compilation? > for example: > #ifdef DHCPCD_PATH > //handling dhcpcd configuration file > #else > //handling dhclient configuration file > #endif > > If this could be achieved, what should follow "#ifdef"? Should I include > any header file? I changed NM a few months ago to allow either DHCP client at runtime, so it's possible that support for *both* is compiled in. It's then selectable via the config file /etc/NetworkManager/NetworkManager.conf or (older name) /etc/NetworkManager/nm-system-settings.conf. I think at this point, you should probably look for that file and use the DHCP client listed there. The actual code that determines which DHCP client to use at run time is in: src/dhcp-manager/nm-dhcp-manager.c::get_client_type() It's probably sufficient to grab the variable out of the .conf file, and if that doesn't exist or isn't valid, use dhclient (if it's installed) and then fallback dhcpcd (if it's installed). Dan ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: [PATCH] Fix suffix for 1900 MHz UMTS
On Fri, 2010-06-04 at 17:18 -0400, David Rochberg wrote: > A little nit: Pushed, thanks. Dan > > --- > introspection/mm-modem-gsm.xml |2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > > diff --git a/introspection/mm-modem-gsm.xml > b/introspection/mm-modem-gsm.xml > index 258ddac..ea23adc 100644 > --- a/introspection/mm-modem-gsm.xml > +++ b/introspection/mm-modem-gsm.xml > @@ -97,7 +97,7 @@ > > WCDMA 3GPP UMTS 1700 MHz (Class > IX) > > - > + > WCDMA 3GPP UMTS 1900 MHz (Class > II) > > > -- > 1.7.0.1 > > > ___ > networkmanager-list mailing list > networkmanager-list@gnome.org > http://mail.gnome.org/mailman/listinfo/networkmanager-list ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
[PATCH] Fix suffix for 1900 MHz UMTS
A little nit: --- introspection/mm-modem-gsm.xml |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/introspection/mm-modem-gsm.xml b/introspection/mm-modem-gsm.xml index 258ddac..ea23adc 100644 --- a/introspection/mm-modem-gsm.xml +++ b/introspection/mm-modem-gsm.xml @@ -97,7 +97,7 @@ WCDMA 3GPP UMTS 1700 MHz (Class IX) - + WCDMA 3GPP UMTS 1900 MHz (Class II) -- 1.7.0.1 ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
UI for dealing with certs appears insecure
I regularly want to access a particular wireless network using a password whose security I care about, but I currently avoid using this network because of security concerns I describe here. This wireless network has what appears to me to be a reasonable authentication mechanism for the desired level of security: it's using WPA2 Enterprise with EAP-MSCHAPv2 authentication (I hope I'm using the right terminology here), and has a valid cert signed by a well-known root CA, for a hostname that makes sense in context (i.e., its domain is the domain of the company operating the wireless network). However, the NetworkManager UI doesn't give me confidence in the handling of the security of my password, since it prompts me for (all at once): * my username * my password * what root CA should be used (if any) to validate the cert In this particular case, it seems somebody could steal my password if they set up a wireless network nearby with the same SSID, a stronger signal, and a valid cert purchased from the same CA (but for a different domain). Or, if I choose the full root cert list for the CA (since I really don't know any other way to figure out what the right root CA is other than finding a friend with a Mac to connect to that wireless network), the attacker could use a valid cert from any CA. It seems to me that in cases where certificates are involved: * the prompt for my username and password should not happen until the cert has been checked, and it should display information about the cert, i.e.: + if the cert was signed by a CA in the root cert list, the hostname the cert is for (and probably the CA that signed it, and perhaps the fingerprint) + otherwise, the cert's fingerprint * when I enter a username and password for such a prompt, it should only be used for wireless with that SSID+Cert combination, and not for other wireless networks with the same SSID. It seems like this would prevent the attack described above, and also improve security in the self-signed cert case. A UI that worked this way would make me comfortable accessing the network in question using NetworkManager. Is this a reasonable request? Are there reasons the current UI is preferable to such a UI? -David -- L. David Baron http://dbaron.org/ Mozilla Corporation http://www.mozilla.com/ ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: Looking for advice about using NM with a Sprint U301 modem
On 06/04/2010 12:01 AM, John Mahoney wrote: I'm looking into the 598 now, and I'm having a little trouble getting NM to recognize it. I plug in the device and I can see by using dmesg that the sierra driver gets loaded, but when I run nm-connection-editor and try to edit the connection, it doesn't get listed in the device dropdown. Is there anything I have to do to get it recognized? I also haven't been able to find any good guides on setting it up. I'm running Ubuntu Karmic right now. On Sierra's site there is a link(http://www.sierrawireless.com/en/Support/knowledgebase.aspx) in the knowledge base which says: "**Note: Ubuntu 9.10 distribution is not supported with Sierra Wireless modems. Ubuntu 9.04 distribution is still supported with all Sierra Wireless modems listed in this KB article. We expect the issue to be fixed in Ubuntu 10.4." If you require Ubuntu 9.10 and Sierra isn't working get a Novatel. -- John I am sure Sierra knows better than I, even so, Ubu 9.10 connected both my 875U & 885U Sierra modems to ATT with minimum effort. Simply boot 9.10 Live CD, see it offer to install modem, click 5 times, then watch NM connect. I gave a few Ubuntu 9.1 CDs to ATT stores for use as trouble shooting tool. Eric ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
How to conditionally compile plug-in source file to only support dhcpcd or dhclient?
Hi Dan, Gentoo NM plugin has to read dhcp hostname and clientid from configuration file of dhcpcd or dhclient. Currently I just test whether the configuration file of dhcpcd or dhclient exists to determine which one is used. Could I make use of the switch "--with-dhcpcd=yes" to enable conditional compilation? for example: #ifdef DHCPCD_PATH //handling dhcpcd configuration file #else //handling dhclient configuration file #endif If this could be achieved, what should follow "#ifdef"? Should I include any header file? -- Cheers, Mu Qiao GnuPG fingerprint: 92B1 B0C4 8D14 F8C4 EFA5 3ACC 30B3 0DE4 17B1 57E9 signature.asc Description: OpenPGP digital signature ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: Failed to build from latest git source: checking for libnl address caching bug... ?
On Friday 04 of June 2010 10:45:03 toabctl wrote: > I tried to build the latest git source with debian squeeze. > > For i386, all works fine. But for armel, i got the following error: > > ./autogen.sh > ... > ... > checking for LIBNL... yes > checking for libnl address caching bug... ? > configure: error: libnl test program failed > > > Any ideas? > > Cheers, > > Tom configure process tries to determine the behaviour of libnl. It uses a C test program. Look at ./m4/libnl-check.m4 Also look at ./config.log to find out what the program returns. It probably returns 3: either "couldn't connect to netlink:" or "couldn't fill address cache:" Jirka ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: Looking for advice about using NM with a Sprint U301 modem
On Fri, Jun 4, 2010 at 01:01, John Mahoney wrote: >> I'm looking into the 598 now, and I'm having a little trouble getting >> NM to recognize it. I plug in the device and I can see by using dmesg >> that the sierra driver gets loaded, but when I run >> nm-connection-editor and try to edit the connection, it doesn't get >> listed in the device dropdown. Is there anything I have to do to get >> it recognized? I also haven't been able to find any good guides on >> setting it up. I'm running Ubuntu Karmic right now. > > On Sierra's site there is a > link(http://www.sierrawireless.com/en/Support/knowledgebase.aspx) in > the knowledge base which says: > > "**Note: Ubuntu 9.10 distribution is not supported with Sierra > Wireless modems. Ubuntu 9.04 distribution is still supported with all > Sierra Wireless modems listed in this KB article. > We expect the issue to be fixed in Ubuntu 10.4." > > If you require Ubuntu 9.10 and Sierra isn't working get a Novatel. > > -- > John > Thanks for the follow up John! So far the 598U seems to be working alright, but I'll keep the novatel in mind if it should prove unreliable. We do plan to upgrade our systems to be running lucid as soon as possible, but they'll be on karmic a little longer yet. ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
RE: [PATCH 2/2] add imei modem property
> -Original Message- > From: Dan Williams [mailto:d...@redhat.com] > Sent: den 2 juni 2010 09:19 > To: Jason Glasgow > Cc: Torgny Johansson; networkmanager-list@gnome.org; David Rochberg > Subject: Re: [PATCH 2/2] add imei modem property > > On Thu, 2010-05-27 at 13:28 -0400, Jason Glasgow wrote: > > I'd like agree with David's suggestion. Adding this > property to the > > Modem object and calling it EquipmentIdentifier will > simplify the use > > case -- allowing clients of the Modem Manager DBUS interface to > > uniquely identify devices with a persistent id independent of modem > > type. > > Yeah, that sounds good to me. The property's spec should state that > (currently) the value is a string which will be either the > IMEI (for GSM-based devices) or (for CDMA) the hex-format ESN/MEID. Ok, I've changed the patch so that the property (on the modem interface) is now called EquipmentIdentity. I've implemented it for GSM-devices but not for CDMA-devices. The new patch is attached and it should be applied _after_ the 0001-Added-UnlockRetries-property-to-the-modem-interface.patch that I sent a few days back. /Torgny > > Dan > > > > > -Jason > > > > > > On Thu, May 27, 2010 at 5:57 AM, Torgny Johansson > > wrote: > > > -Original Message- > > > From: rochb...@google.com [mailto:rochb...@google.com] On > > > Behalf Of David Rochberg > > > Sent: den 26 maj 2010 15:31 > > > To: Torgny Johansson > > > Cc: Dan Williams; networkmanager-list@gnome.org > > > Subject: Re: [PATCH 2/2] add imei modem property > > > > > > > > For things like this with similar semantics but different > > > names, would it make sense to hang them off of modem and > > > define the semantics according to the protocol? > That is, a > > > property like EquipmentIdentifier, with GSM and > CDMA phones > > > supplying IMEI and MEID, respectively. > > > > > > I've moved the property to gsm card but I'll wait until this > > has been discussed before submitting it. > > > > To me your suggestion sounds good. Which way do you > want to go > > Dan? > > > > /Torgny > > > > > > > > > > -david > > > > > > > > > On Wed, May 26, 2010 at 3:38 AM, Torgny Johansson > > > wrote: > > > > > > > > > onsdagen den 26 maj 2010 07.57.10 skrev > Dan Williams: > > > > > > > On Tue, 2010-05-25 at 10:42 +0200, Torgny > Johansson > > wrote: > > > > > Hi, > > > > > > > > > > this patch adds the IMEI of the modem as a > > property > > > so that it is > > > > > available at an early stage to simplify the > > connect > > > sequence. > > > > > > > > So this can't be a property on the base > modem object > > > because CDMA > > > > devices don't have an IMEI (well, not > entirely true, > > > but mostly true) > > > > but instead have an MEID. I'd rather have the > > property on the > > > > org.freedesktop.ModemManager.Gsm.Card interface > > > (mm-modem-gsm-card.c for > > > > the property's definition) instead and > overridden by > > > the generic GSM > > > > class. Does that sound OK? > > > > > > > > > Ah, makes sense. I'll try to move it. > > > > > > Thanks! > > > > > > /Torgny > > > > > > > > > > > > > > Dan > > > > > > > > > Regards > > > > > Torgny Johansson > > > > > > > > > > --- > > > > > > > > > > diff --git a/introspection/mm-modem.xml > > > b/introspection/mm-modem.xml > > > > > index 09512f8..8eff674 100644 > > > > > --- a/introspection/mm-modem.xml > > > > > +++ b/introspection/mm-modem.xml > > > > > @@ -113,6 +113,12 @@ > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > +access="read"> > > > > > + > > > > > +The IMEI of the device. > > > > > + > > > > > + > > > > > + > > > > > > > > > > > access="read"> > > > > > > > > > > > > > > > > > > > > Empty if the device is usable > without an > > > unlock code or has > > > > > already > > > > > > > > > > diff --git a/src/mm-generic-gsm.c > > b/src/mm-generic-gsm.c > >
Failed to build from latest git source: checking for libnl address caching bug... ?
I tried to build the latest git source with debian squeeze. For i386, all works fine. But for armel, i got the following error: ./autogen.sh ... ... checking for LIBNL... yes checking for libnl address caching bug... ? configure: error: libnl test program failed Any ideas? Cheers, Tom ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
Re: What does the NM Applet button do?
On Friday 04 of June 2010 02:08:46 James K. Williams wrote: > When you click left on the NetworkManager applet, there is a list of > connections, and you must select one of the radio buttons there to get > networking working. What exactly does selecting that radio button do? > > Here's why I ask. I don't really need NM at all, because I can edit all the > network configuration files by hand and then restart network services with > > # services network restart > > This should be enough to bring networking up, but it's not: I still have to > select one of those radio buttons in the NM applet. To try to determine > what pressing that button does, I have checked services before and after, > and there is no difference. Also, I have checked the config files before > and after, and they are not changed. So what happens under the hood when I > select one of the buttons in the applet? Why doesn't networking come back > on until I press it? > > I'm using NetworkManager 0.7.1 on Fedora 9. > > Thanks. There are two services to manage networks: network - legacy service used on Fedora NetworkManager - meant to replace network service You want probably use just one of them. So to use 'network' service, stop NetworkManager: service NetworkManager stop and disable it not to be run any more: chkconfig NetworkManager off Then you can manually configure ifcfg- files and do ifup/ifdown. NM applet is a client for NetworkManager daemon. When you click an item in NM applet, it communicates with NetworkManager daemon via D-Bus to activate the network connection. Jirka ___ networkmanager-list mailing list networkmanager-list@gnome.org http://mail.gnome.org/mailman/listinfo/networkmanager-list
