[wily-proposed only] NetworkManager crashed with SIGSEGV in ofono_enumerate_devices_done()

Bug #1492126 reported by Martin Pitt
14
This bug affects 2 people
Affects Status Importance Assigned to Milestone
network-manager (Ubuntu)
Fix Released
Medium
Mathieu Trudel-Lapierre

Bug Description

This can be trivially reproduced by installing the wily-proposed network-manager on i386 and rebooting. This is also spotted by systemd's autopkgtest (http://autopkgtest.ubuntu.com/packages/s/systemd/wily/i386/).

In the journal I see

Sep 04 08:32:15 autopkgtest NetworkManager[718]: (NetworkManager:718): GLib-GObject-WARNING **: /build/glib2.0-M4BESt/glib2.0-2.45.6/./gobject/gsignal.c:2506: signal 'ModemAdded' is invalid for instance '0x997d760' of type 'GDBusProxy'
Sep 04 08:32:15 autopkgtest NetworkManager[718]: (NetworkManager:718): GLib-GObject-WARNING **: /build/glib2.0-M4BESt/glib2.0-2.45.6/./gobject/gsignal.c:2506: signal 'ModemRemoved' is invalid for instance '0x997d760' of type 'GDBusProxy'
Sep 04 08:32:15 autopkgtest NetworkManager[718]: (NetworkManager:718): GLib-CRITICAL **: g_variant_get_va: assertion 'value != NULL' failed
Sep 04 08:32:15 autopkgtest systemd[1]: NetworkManager.service: Main process exited, code=dumped, status=11/SEGV

I suppose the g_critical is not the thing it fails on, as that would be a SIGABRT.

ProblemType: Crash
DistroRelease: Ubuntu 15.10
Package: network-manager 1.0.4-0ubuntu2
ProcVersionSignature: Ubuntu 4.2.0-7.7-generic 4.2.0
Uname: Linux 4.2.0-7-generic i686
ApportVersion: 2.18-0ubuntu9
Architecture: i386
Date: Fri Sep 4 08:40:19 2015
ExecutablePath: /usr/sbin/NetworkManager
ExecutableTimestamp: 1440792479
IpRoute:
 default via 10.0.2.2 dev eth0
 10.0.2.0/24 dev eth0 proto kernel scope link src 10.0.2.15
IwConfig: Error: [Errno 2] No such file or directory: 'iwconfig'
NetworkManager.state:
 [main]
 NetworkingEnabled=true
 WirelessEnabled=true
 WWANEnabled=true
 WimaxEnabled=true
ProcCmdline: /usr/sbin/NetworkManager --no-daemon
ProcCwd: /
ProcEnviron:
 LANG=en_US.UTF-8
 PATH=(custom, no user)
RfKill: Error: [Errno 2] No such file or directory: 'rfkill'
SegvAnalysis:
 Segfault happened at: 0xb701f259 <g_variant_iter_loop+25>: mov 0xc(%esi),%edi
 PC (0xb701f259) ok
 source "0xc(%esi)" (0x0000000c) not located in a known VMA region (needed readable region)!
 destination "%edi" ok
SegvReason: reading NULL VMA
Signal: 11
SourcePackage: network-manager
StacktraceTop:
 g_variant_iter_loop () from /lib/i386-linux-gnu/libglib-2.0.so.0
 ?? () from /usr/lib/i386-linux-gnu/NetworkManager/libnm-wwan.so
 ?? () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
 ?? () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
 ?? () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
Title: NetworkManager crashed with SIGSEGV in g_variant_iter_loop()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups:

nmcli-con: Error: command ['nmcli', '-f', 'all', 'con'] failed with exit code 8: Error: NetworkManager is not running.
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
Martin Pitt (pitti) wrote :
information type: Private → Public
tags: added: block-proposed
Changed in network-manager (Ubuntu):
assignee: nobody → Mathieu Trudel-Lapierre (mathieu-tl)
summary: - NetworkManager crashed with SIGSEGV in g_variant_iter_loop()
+ [wily-proposed only] NetworkManager crashed with SIGSEGV in
+ g_variant_iter_loop()
Changed in network-manager (Ubuntu):
importance: Undecided → Medium
Revision history for this message
Martin Pitt (pitti) wrote : Re: [wily-proposed only] NetworkManager crashed with SIGSEGV in g_variant_iter_loop()

Importance medium as this only affects wily-proposed, and failing tests kept this from promoting. But I added block-proposed anyway as you were disabling most NM autopkgtests.

Revision history for this message
Apport retracing service (apport) wrote :

StacktraceTop:
 g_variant_iter_loop (iter=0x0, format_string=0xb424daba "(&oa{sv})") at /build/glib2.0-M4BESt/glib2.0-2.45.6/./glib/gvariant.c:5616
 ofono_enumerate_devices_done (proxy=0x8c91760, res=0x8cbf810, user_data=0x8caa708) at nm-modem-manager.c:329
 g_task_return_now (task=task@entry=0x8cbf810) at /build/glib2.0-M4BESt/glib2.0-2.45.6/./gio/gtask.c:1104
 g_task_return (task=0x8cbf810, type=<optimized out>) at /build/glib2.0-M4BESt/glib2.0-2.45.6/./gio/gtask.c:1162
 reply_cb (connection=0x8c9c818, res=0x8cbf888, user_data=0x8cbf810) at /build/glib2.0-M4BESt/glib2.0-2.45.6/./gio/gdbusproxy.c:2569

Revision history for this message
Apport retracing service (apport) wrote : Stacktrace.txt
Revision history for this message
Apport retracing service (apport) wrote : StacktraceSource.txt
Revision history for this message
Apport retracing service (apport) wrote : ThreadStacktrace.txt
tags: removed: need-i386-retrace
Revision history for this message
Martin Pitt (pitti) wrote :

Adjusting bug title to make this more obvious. Apparently ofono_enumerate_devices_done() calls g_variant_iter_loop() on a NULL iterator.

summary: [wily-proposed only] NetworkManager crashed with SIGSEGV in
- g_variant_iter_loop()
+ ofono_enumerate_devices_done()
Revision history for this message
dino99 (9d9) wrote :

also a show stopper:
 not all the network-manager packages have been upgraded to 1.0.4-0ubuntu2, some are still 0.9.10

reported as lp:1489154

Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in network-manager (Ubuntu):
status: New → Confirmed
Revision history for this message
dino99 (9d9) wrote :

Also from journalctl on Wily with nm 0.9.10:

 NetworkManager[806]: <warn> SCPlugin-Ofono: file doesn't exist: /var/lib/ofono

ofono is not installed on that desktop which does not need it anyway

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

dino99, this is unrelated. Packages won't half-migrate from proposed unless one actually enables the proposed repository on their system and manually picks packages to install... This would explicitly be a very Bad thing to do, and strongly discouraged.

In Progress -- I'm looking at that crash right now.

Changed in network-manager (Ubuntu):
status: Confirmed → In Progress
Revision history for this message
Martin Pitt (pitti) wrote :

The crash got fixed, thanks Mathieu! Removing block-proposed tag.

tags: removed: block-proposed
Changed in network-manager (Ubuntu):
status: In Progress → Fix Committed
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package network-manager - 1.0.4-0ubuntu3

---------------
network-manager (1.0.4-0ubuntu3) wily; urgency=medium

  * New upstream NM release post-feature-freeze. (LP: #1493934)
    (previous revisions never transitioned from proposed)

  [ Martin Pitt ]
  * debian/tests/nm: Revert the previous disabling of tests, they are all
    conceptually still valid.
  * debian/tests/nm: Reduce timeouts to something reasonable; waiting 3
    minutes for detecing an AP should not count as "success", and it takes too
    long for tests to fail.
  * debian/tests/nm: Fix ColdplugWifi.test_no_ap: with 1.0 the state is now
    "UNAVAILABLE" instead of "DISCONNECTED"; so check for either.

  [ Mathieu Trudel-Lapierre ]
  * debian/patches/git_config_parse_cmdline_args_8a22a981.patch: don't shadow
    the config_cli parameter for early config with a local copy. (LP: #1492168)
  * d/p/0001-wwan-add-support-for-using-oFono-as-a-modem-manager.patch: fix
    signal handling for oFono using GDBus; it's a little simpler now, watching
    for signals in general and matching only ModemAdded/ModemRemoved. Inspired
    from urfkill code. (LP: #1492126)
  * debian/patches/git_fix_duplicate_unittest_names_7b70a840.patch: replace
    duplicate unit test names, which is a new restriction set by GLib.

 -- Mathieu Trudel-Lapierre <email address hidden> Thu, 10 Sep 2015 14:53:16 -0400

Changed in network-manager (Ubuntu):
status: Fix Committed → Fix Released
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.