Bug 732845

Summary: systemd dies during yum upgrades
Product: [Fedora] Fedora Reporter: Michal Jaegermann <michal>
Component: systemdAssignee: Lennart Poettering <lpoetter>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: rawhideCC: harald, johannbg, kay, lpoetter, metherid, mschmidt, notting, plautrba
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-08-24 07:56:13 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
some gdb information from a core file produced by systemd none

Description Michal Jaegermann 2011-08-23 21:17:53 UTC 
Description of problem:

After yum finished updating a number of packages on a system any attempt to 'reboot', 'halt', 'poweroff', or any other command of that sort end up with the following response:

Failed to open /dev/initctl: No such device or address
Failed to talk to init daemon.

/dev/initctl actually does exist, making this even more confusing, but checking processes reveals that systemd is gone.  Moreover in log files it is possible to find something of that kind:

yum[1759]: Updated: ImageMagick-c++-devel-6.7.1.9-1.fc17.x86_64
systemd[1]: Reloading.
systemd[1]: Caught <ABRT>, dumped core as pid 2166.
systemd[1]: Freezing execution.
yum[1759]: Updated: lvm2-2.02.88-1.fc17.x86_64

abrtd failed to catch that core.  With systemd-debuginfo and dbus-debuginfo installed gdb shows that dbus libraries call 
_dbus_warn_check_failed () with format=0x3c33a36010 "arguments to %s() were incorrect, assertion \"%s\" failed in file %s line %d.\nThis is normally a bug in some application using the D-Bus library.\n"
What values are supposed to be passed here to this format is for me unclear.
More of gdb output is attached. A core file is 5.6M in size (but xz reduces that  to 198K and that can be available if requested).

Bombing out from init does not look like an acceptable action does not matter what other userspace is doing.

Version-Release number of selected component (if applicable):
systemd-33-1.fc16

How reproducible:
This is the second incident of the same kind.  Both were with 
systemd-33-1.fc16.  I do not know if this is significant but the previous instance of "systemd[1]: Caught <ABRT>, dumped core ..." was followed immediately by "yum[1360]: Updated: lvm2-2.02.87-1.fc17.x86_64" so it was an lvm2 update as well.

Actual results:
Non-rebootable broken system.

Expected results:
Running init; bombing out from init does not look like an acceptable action does not matter what other userspace is doing.

Additional info:
This is the list of update which were running around time when this ABRT happened:

Aug 23 12:45:00 Updated: policycoreutils-2.1.4-1.fc17.x86_64
Aug 23 12:45:03 Updated: 6:kdelibs-common-4.7.0-4.fc17.x86_64
Aug 23 12:45:04 Updated: kde-settings-4.7-5.fc17.noarch
Aug 23 12:45:16 Updated: 6:kdelibs-4.7.0-4.fc17.x86_64
Aug 23 12:45:17 Updated: kdebase-runtime-flags-4.7.0-5.fc17.noarch
Aug 23 12:45:24 Updated: kdebase-runtime-4.7.0-5.fc17.x86_64
Aug 23 12:45:26 Updated: kdebase-runtime-libs-4.7.0-5.fc17.x86_64
Aug 23 12:45:26 Updated: ksysguardd-4.7.0-8.fc17.x86_64
Aug 23 12:45:28 Updated: kdebase-workspace-libs-4.7.0-8.fc17.x86_64
Aug 23 12:45:43 Updated: kdebase-workspace-4.7.0-8.fc17.x86_64
Aug 23 12:45:43 Updated: emacs-librep-0.92.1-1.fc17.x86_64
Aug 23 12:45:45 Updated: kbd-misc-1.15.3-1.fc17.noarch
Aug 23 12:45:46 Updated: kbd-1.15.3-1.fc17.x86_64
Aug 23 12:45:47 Updated: p11-kit-0.4-1.fc17.x86_64
Aug 23 12:45:49 Updated: p11-kit-devel-0.4-1.fc17.x86_64
Aug 23 12:45:51 Updated: pm-utils-1.4.1-12.fc17.x86_64
Aug 23 12:45:51 Updated: emacs-librep-el-0.92.1-1.fc17.x86_64
Aug 23 12:45:52 Updated: system-config-printer-1.3.6-1.fc17.x86_64
Aug 23 12:45:53 Updated: policycoreutils-python-2.1.4-1.fc17.x86_64
Aug 23 12:45:54 Updated: ImageMagick-c++-devel-6.7.1.9-1.fc17.x86_64
  <ABRT here >
Aug 23 12:45:56 Updated: lvm2-2.02.88-1.fc17.x86_64  
Aug 23 12:46:08 Updated: rhythmbox-2.90.1-13.git20110822.fc17.x86_64

and this is a similar list for the previous incident:

Aug 14 17:25:16 Updated: device-mapper-1.02.66-1.fc17.x86_64
Aug 14 17:25:18 Updated: device-mapper-libs-1.02.66-1.fc17.x86_64
Aug 14 17:25:20 Updated: 2:gimp-libs-2.6.11-21.fc17.x86_64
Aug 14 17:25:21 Updated: device-mapper-event-libs-1.02.66-1.fc17.x86_64
Aug 14 17:25:30 Updated: 2:gimp-2.6.11-21.fc17.x86_64
Aug 14 17:25:31 Updated: 12:dhcp-libs-4.2.2-2.fc17.x86_64
Aug 14 17:25:33 Updated: ufraw-common-0.18-3.fc17.x86_64
Aug 14 17:25:33 Updated: 12:dhcp-common-4.2.2-2.fc17.x86_64
Aug 14 17:25:37 Updated: device-mapper-event-1.02.66-1.fc17.x86_64
Aug 14 17:25:38 Updated: lvm2-libs-2.02.87-1.fc17.x86_64
Aug 14 17:25:38 Updated: 2:gimp-devel-tools-2.6.11-21.fc17.x86_64
Aug 14 17:25:42 Updated: 2:gimp-devel-2.6.11-21.fc17.x86_64
Aug 14 17:25:42 Updated: ksysguardd-4.7.0-5.fc17.x86_64
Aug 14 17:25:44 Updated: kdebase-workspace-libs-4.7.0-5.fc17.x86_64
Aug 14 17:25:55 Updated: kdebase-workspace-4.7.0-5.fc17.x86_64
  <ABRT here >
Aug 14 17:25:57 Updated: lvm2-2.02.87-1.fc17.x86_64
Aug 14 17:25:59 Updated: 12:dhclient-4.2.2-2.fc17.x86_64
Comment 1 Michal Jaegermann 2011-08-23 21:19:44 UTC 
Created attachment 519523 [details]
some gdb information from a core file produced by systemd
Comment 2 Harald Hoyer 2011-08-24 07:56:13 UTC 

*** This bug has been marked as a duplicate of bug 732020 ***