Revert latest nm

[Sergio Schvezov]

Since the latest nm, everything is completely broken...

On desktop, when on ethertet and wifi, the applet displays nothing as connected; on the phone, no route is set, ifconfig fails badly and networking is completely broken as seen here:

phablet@ubuntu-phablet:~$ ifconfig
hD���
�: error fetching interface information: Device not found
phablet@ubuntu-phablet:~$ netstat -rn
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
10.225.0.0 0.0.0.0 255.255.0.0 U 0 0 0 wlan1
phablet@ubuntu-phablet:~$ nmcli d
DEVICE TYPE STATE CONNECTION
ccmni1 ethernet connected ccmni1
wlan1 wifi connected vip.latam
ril_0 gsm disconnected --
ccmni0 ethernet unmanaged --
ccmni2 ethernet unmanaged --
ifb0 ifb unmanaged --
ifb1 ifb unmanaged --
ip6tnl0 ip6tnl unmanaged --
tunl0 ipip unmanaged --
lo loopback unmanaged --
sit0 sit unmanaged --

Please fix or revert!

ProblemType: Bug
DistroRelease: Ubuntu 15.04
Package: network-manager 0.9.10.0-4ubuntu6
ProcVersionSignature: Ubuntu 3.18.0-13.14-generic 3.18.5
Uname: Linux 3.18.0-13-generic x86_64
ApportVersion: 2.16.1-0ubuntu2
Architecture: amd64
CurrentDesktop: Unity
Date: Tue Feb 17 19:48:41 2015
IfupdownConfig:
 # interfaces(5) file used by ifup(8) and ifdown(8)
 auto lo
 iface lo inet loopback
InstallationDate: Installed on 2014-10-25 (115 days ago)
InstallationMedia: Ubuntu 14.10 "Utopic Unicorn" - Release amd64 (20141022.1)
IpRoute:
 default via 10.225.220.1 dev wlan0 proto static metric 1024
 10.0.3.0/24 dev lxcbr0 proto kernel scope link src 10.0.3.1
 10.225.0.0/16 dev wlan0 proto kernel scope link src 10.225.220.59
 169.254.0.0/16 dev lxcbr0 scope link metric 1000
 192.168.122.0/24 dev virbr0 proto kernel scope link src 192.168.122.1
NetworkManager.state:
 [main]
 NetworkingEnabled=true
 WirelessEnabled=true
 WWANEnabled=true
 WimaxEnabled=true
RfKill:
 0: phy0: Wireless LAN
  Soft blocked: no
  Hard blocked: no
SourcePackage: network-manager
UpgradeStatus: Upgraded to vivid on 2014-11-17 (92 days ago)
modified.conffile..etc.NetworkManager.NetworkManager.conf: [modified]
mtime.conffile..etc.NetworkManager.NetworkManager.conf: 2014-11-11T12:46:14.739895
nmcli-nm: Error: command ['nmcli', '-f', 'all', 'nm'] failed with exit code 2: Error: Object 'nm' is unknown, try 'nmcli help'.

[Sergio Schvezov (sergiusens)]

If I shutdown the ofono context I get

phablet@ubuntu-phablet:~$ ifconfig
ccmni1 Link encap:Ethernet HWaddr 0a:12:fe:4b:6d:83
          UP RUNNING NOARP MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:31 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B) TX bytes:3298 (3.2 KB)

hD���
�: error fetching interface information: Device not found

[Mathieu Trudel-Lapierre (cyphermox)]

Except NetworkManager itself doesn't make any new interfaces available to you, and much less available of ifconfig.

As for the ccmni0 and the state of the ril device, that looks like something that was fixed recently, but you seem to be running the most recent revision (though it should be confirmed with the version on the phone).

Please attach syslog for the touch install so we can see what happens in why devices are showing up the way they are.

[Tony Espy (awe)]

A couple of comments...

First, this bug is overly broad... The indicator on the phone does work when WiFi or mobile data are connected, however there are some other issues.

Regarding problems on touch. Standard procedure is to include device names and image #s. Also, the indicator does work when WiFi or mobile data are connected, which is different than what you describe on the desktop.

Regarding the ifconfig errors, I have seen this error on krillin and arale running the latest vivid-devel. I haven't checked mako. My original thought was this was a kernel error, however if you're seeing this on the desktop, it may be an issue with ifconfig itself ( part of the net-tools package ).

My touch testing so far has shown similar errors on RTM and vivid-devel with indicator-network icon latency ( see Bug #1339792 ), DNS taking too long ( 10s - 1m ) to be re-established when switching connections, and routing table errors on arale and krillin ( vivid-devel-only ). Of these, only the last seems related to the NM upgrade as the first two problems occur on RTM as well.

I've been tasked with focusing on networking debug this coming week.

As I'm not running vivid on my laptop, perhaps we keep this bug for the desktop problem and try and determine what's actually going on before we decide to cut & run? "Revert latest nm" is not a bug, it's a suggested action and without first determining the root cause, is a bit drastic a suggestion at this time...

[Bryan Quigley (bryanquigley)]

This bug was marked Incomplete some time ago and was supposed to autoclose after 90 days. I'm going to close it manually due to no activity (and likely the issue has gone away/been fixed).