Today's network-manager update completely breaks network-manager service

Bug #1415702 reported by 3vi1
38
This bug affects 8 people
Affects Status Importance Assigned to Milestone
network-manager (Ubuntu)
Triaged
High
Mathieu Trudel-Lapierre

Bug Description

WORKAROUND:

- Remove any phone device paired in bluetooth if it's not needed
OR
- Stop the ofono service with 'sudo service ofono stop'

Removing the phone pairing will let NetworkManager run succesfully, but will make bluetooth DUN/PAN unavailable. Furthermore, it will also disable file sharing options over bluetooth for that device and any other bluetooth features for that device.

Stopping the ofono service will disable modems that require to be used via ofono, but will most modem devices on computers usually use ModemManager instead, and will continue to work adequately.

------------------------------------------------------------------

The network-manager updates in vivid today [network-manager:amd64 (0.9.8.8-0ubuntu34, 0.9.10.0-4ubuntu2)] have completely broken the network-manager service on my system.

--------------
evil@clevo:~$ sudo service network-manager restart
stop: Unknown instance:
network-manager start/running, process 16012
evil@clevo:~$ dmesg | tail
[ 2670.516007] init: network-manager main process ended, respawning
[ 2670.666232] traps: NetworkManager[16326] trap int3 ip:7fa7b9aaef90 sp:7fff3aa42e10 error:0
[ 2670.757044] init: network-manager main process (16326) killed by TRAP signal
[ 2670.757057] init: network-manager main process ended, respawning
[ 2670.898315] traps: NetworkManager[16469] trap int3 ip:7fa47348ff90 sp:7ffffeb81760 error:0
[ 2670.991490] init: network-manager main process (16469) killed by TRAP signal
[ 2670.991502] init: network-manager main process ended, respawning
[ 2671.131611] traps: NetworkManager[16602] trap int3 ip:7f06f28eaf90 sp:7fffefdc5d70 error:0
[ 2671.253569] init: network-manager main process (16602) killed by TRAP signal
[ 2671.253581] init: network-manager main process ended, respawning
...
[ 2672.236202] init: network-manager respawning too fast, stopped
-------------------

I had to manually add eth0 to /etc/network/interfaces and start it outside of network manager as a workaround.

ProblemType: Bug
DistroRelease: Ubuntu 15.04
Package: network-manager 0.9.10.0-4ubuntu2
ProcVersionSignature: Ubuntu 3.18.0-11.12-generic 3.18.3
Uname: Linux 3.18.0-11-generic x86_64
ApportVersion: 2.15.1-0ubuntu4
Architecture: amd64
CurrentDesktop: KDE
Date: Wed Jan 28 20:05:59 2015
IpRoute:
 default via 192.168.1.1 dev eth0
 10.0.3.0/24 dev lxcbr0 proto kernel scope link src 10.0.3.1
 169.254.0.0/16 dev eth0 scope link metric 1000
 192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.154
 192.168.122.0/24 dev virbr0 proto kernel scope link src 192.168.122.1
SourcePackage: network-manager
UpgradeStatus: No upgrade log present (probably fresh install)
modified.conffile..etc.NetworkManager.NetworkManager.conf: [modified]
mtime.conffile..etc.NetworkManager.NetworkManager.conf: 2015-01-25T17:12:21.424810
nmcli-con: Error: command ['nmcli', '-f', 'all', 'con'] failed with exit code 9: Error: nmcli (0.9.10.0) and NetworkManager (unknown) versions don't match. Force execution using --nocheck, but the results are unpredictable.
nmcli-dev: Error: command ['nmcli', '-f', 'all', 'dev'] failed with exit code 8: Error: NetworkManager is not running.
nmcli-nm: Error: command ['nmcli', '-f', 'all', 'nm'] failed with exit code 2: Error: Object 'nm' is unknown, try 'nmcli help'.

Revision history for this message
3vi1 (launchpad-net-eternaldusk) wrote :
Revision history for this message
Mathieu Trudel-Lapierre (cyphermox) wrote :

This appears to be a crash; could you please check /var/crash for a file named _usr_sbin_NetworkManager.0.crash or something similar, and run "ubuntu-bug <file>". This will automatically open a new bug with the crash details and all the necessary information. Thanks!

Revision history for this message
Mathieu Trudel-Lapierre (cyphermox) wrote :

There's enough information on this bug report to actually do something useful:

Jan 28 19:33:14 clevo NetworkManager[27189]: <warn> (hfp/2CD05A55C693_0C3E9FAE619C): failed to look up interface index
Jan 28 19:33:14 clevo NetworkManager[27189]: <info> (hfp/2CD05A55C693_0C3E9FAE619C): device state change: unmanaged -> unavailable (reason 'none') [10 20 0]
Jan 28 19:33:14 clevo NetworkManager[27189]: <error> [1422495194.170276] [NetworkManagerUtils.c:1252] ASSERT_VALID_PATH_COMPONENT(): Failed asserting path component: NULL
Jan 28 19:33:14 clevo NetworkManager[27189]: FATAL: Failed asserting path component: hfp/2CD05A55C693_0C3E9FAE619C
Jan 28 19:33:14 clevo kernel: [ 432.098680] do_trap: 1 callbacks suppressed
Jan 28 19:33:14 clevo kernel: [ 432.098688] traps: NetworkManager[27189] trap int3 ip:7f42584edf90 sp:7fff1dcb1610 error:0

I'll keep the bug open as a master bug for this occurance; but please still try to file the bug report from the crash dump if you can in case there is more to it.

Workaround is to remove any phone device paired in bluetooth if it's not needed. Alternatively, stopping the ofono service should also let NetworkManager run properly.

description: updated
Changed in network-manager (Ubuntu):
status: New → Triaged
importance: Undecided → High
assignee: nobody → Mathieu Trudel-Lapierre (mathieu-tl)
Revision history for this message
3vi1 (launchpad-net-eternaldusk) wrote :

When I try to use ubuntu-bug on the crash file (with and without sudo), apport seems to silently crash without ever getting to the point of uploading the bug.

You were correct in your workaround. I removed the iPhone I had in the bluetooth configuration (as well as some other MAC-named entry I had not seen before), and Network Manager was able to start successfully.

Thank you very much for spending time on this.

Revision history for this message
3vi1 (launchpad-net-eternaldusk) wrote :

I've tried apport on the crash file repeatedly, but I don't see anything after "continue"ing on the first dialog.

There's also nothing in the apport, dmesg, or syslog files (nor command line) to indicate apport is having a problem. The apport service is still running fine (and I've restarted it just in case)... it just doesn't do anything further with the crash file.

Revision history for this message
Mathieu Trudel-Lapierre (cyphermox) wrote :

We've fixed this as bug 1416635; so I'll mark this bug as a duplicate of it now, thanks!

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.