Qemu virt-manager Segmentation fault

Bug #1838312 reported by Hans Peter
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
virt-manager (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

Hi!

I installed all these packages:

sudo apt install qemu
sudo apt install ipxe-qemu-256k-compat-efi-roms libspice-server1 libbluetooth3
sudo apt install libbrlapi0.6 libcacard0 libfdt1 libusbredirparser1 libvirglrenderer0 libxen-4.9 libxenstore3.0
sudo apt install cpu-checker ibverbs-providers ipxe-qemu libibverbs1 libiscsi7 libnl-route-3-200 librados2 librbd1 librdmacm1 msr-tools sharutils
sudo apt install qemu-block-extra qemu-system-common qemu-system-data qemu-system-gui qemu-utils
sudo apt install --no-install-recommends qemu-kvm qemu-system-x86
sudo apt install libauparse0 ebtables gir1.2-gtk-vnc-2.0 gir1.2-libosinfo-1.0 gir1.2-libvirt-glib-1.0 gir1.2-spiceclientglib-2.0 gir1.2-spiceclientgtk-3.0 libvde0 libvdeplug2 libgovirt-common libgovirt2 libgtk-vnc-2.0-0 libgvnc-1.0-0 libosinfo-1.0-0 libphodav-2.0-0 libphodav-2.0-common libspice-client-glib-2.0-8 libspice-client-gtk-3.0-5 libusbredirhost1 libvirt-clients libvirt-daemon libvirt-daemon-driver-storage-rbd libvirt-daemon-system libvirt-glib-1.0-0 libvirt0 osinfo-db python3-libvirt python3-libxml2 spice-client-glib-usb-acl-helper vde2 vde2-cryptcab virt-viewer virtinst virt-manager

without the i386 packages for Qemu because I want only 64 bit.

I installed all these packages without error, but when I run

# virt-manager

Output: ...shows me:

Segmentation fault

My hardware is 100% ok.
Maybee a broken lib?

How can I fix that?

Revision history for this message
Hans Peter (green-world78) wrote :

syslog:

kernel: [ 2003.888116] virt-manager[16014]: segfault at 32d0 ip 00000000000032d0 sp 00007ffeb09ac658 error 14 in python3.7[400000+21000]

kernel: [ 2003.888124] Code: Bad RIP value.

Changed in qemu:
status: New → Invalid
Revision history for this message
Christian Ehrhardt  (paelzer) wrote :

That is a long list of install commands Hans Peter, wouldn't dependencies just take care of it as well?
Anyway, I installed the same set of packages and it works fine for me.

Which Ubuntu release are you on?
What machine are you on?
Any further configuration done to libvirt yet?
Maybe a list of all installed packages and their versions (dpkg -l)?

You might use apport-collect [1] to gather some of that info automatically.

[1]: http://manpages.ubuntu.com/manpages/bionic/man1/apport-bug.1.html

Changed in virt-manager (Ubuntu):
status: New → Incomplete
Revision history for this message
Hans Peter (green-world78) wrote :

@ Christian Ehrhardt,

I'm on Ubuntu Mate 19.04.

Machine: AMD, 64 bit

I done no configuration of 'libvirt' yet.

I don't wanna install apport.

I add installed package-list at attachment.

I hope it will helps.

Revision history for this message
Hans Peter (green-world78) wrote :
Download full text (45.1 KiB)

I found crash-report from python 3.7 similiar to this error:

kernel: [ 2003.888116] virt-manager[16014]: segfault at 32d0 ip 00000000000032d0 sp 00007ffeb09ac658 error 14 in python3.7[400000+21000]

kernel: [ 2003.888124] Code: Bad RIP value.

look here:

ProblemType: Crash
Date: Fri Jul 26 15:03:11 2019
ExecutablePath: /usr/bin/session-installer
ExecutableTimestamp: 1458142842
InterpreterPath: /usr/bin/python3.7
ProcCmdline: /usr/bin/python3 /usr/bin/session-installer
ProcCwd: /
ProcEnviron:
 SHELL=/bin/bash
 LC_NAME=en_US.UTF-8
 LC_MEASUREMENT=en_US.UTF-8
 LC_PAPER=en_US.UTF-8
 LC_MONETARY=en_US.UTF-8
 LC_NUMERIC=en_US.UTF-8
 LC_TIME=en_US.UTF-8
 LC_IDENTIFICATION=en_US.UTF-8
 LANGUAGE=en_US
 LC_TELEPHONE=de_DE.UTF-8
 LC_ADDRESS=de_DE.UTF-8
 LANG=en_US.UTF-8
ProcMaps:
 00400000-00421000 r--p 00000000 fd:01 2621992 /usr/bin/python3.7
 00421000-00655000 r-xp 00021000 fd:01 2621992 /usr/bin/python3.7
 00655000-00801000 r--p 00255000 fd:01 2621992 /usr/bin/python3.7
 00801000-00802000 r--p 00400000 fd:01 2621992 /usr/bin/python3.7
 00802000-008a8000 rw-p 00401000 fd:01 2621992 /usr/bin/python3.7
 008a8000-008cb000 rw-p 00000000 00:00 0
 02651000-0296c000 rw-p 00000000 00:00 0 [heap]
 7f0869a78000-7f0869c78000 rw-p 00000000 00:00 0
 7f0869c78000-7f0869c83000 r--p 00000000 fd:01 11010294 /lib/x86_64-linux-gnu/libdbus-1.so.3.19.9
 7f0869c83000-7f0869cb0000 r-xp 0000b000 fd:01 11010294 /lib/x86_64-linux-gnu/libdbus-1.so.3.19.9
 7f0869cb0000-7f0869cc5000 r--p 00038000 fd:01 11010294 /lib/x86_64-linux-gnu/libdbus-1.so.3.19.9
 7f0869cc5000-7f0869cc6000 r--p 0004c000 fd:01 11010294 /lib/x86_64-linux-gnu/libdbus-1.so.3.19.9
 7f0869cc6000-7f0869cc7000 rw-p 0004d000 fd:01 11010294 /lib/x86_64-linux-gnu/libdbus-1.so.3.19.9
 7f0869cd8000-7f0869cda000 r--p 00000000 fd:01 2760872 /usr/lib/python3.7/lib-dynload/_json.cpython-37m-x86_64-linux-gnu.so
 7f0869cda000-7f0869ce5000 r-xp 00002000 fd:01 2760872 /usr/lib/python3.7/lib-dynload/_json.cpython-37m-x86_64-linux-gnu.so
 7f0869ce5000-7f0869ce7000 r--p 0000d000 fd:01 2760872 /usr/lib/python3.7/lib-dynload/_json.cpython-37m-x86_64-linux-gnu.so
 7f0869ce7000-7f0869ce8000 r--p 0000e000 fd:01 2760872 /usr/lib/python3.7/lib-dynload/_json.cpython-37m-x86_64-linux-gnu.so
 7f0869ce8000-7f0869ce9000 rw-p 0000f000 fd:01 2760872 /usr/lib/python3.7/lib-dynload/_json.cpython-37m-x86_64-linux-gnu.so
 7f0869cf0000-7f0869cf4000 r--p 00000000 fd:01 2624107 /usr/lib/x86_64-linux-gnu/libapt-inst.so.2.0.0
 7f0869cf4000-7f0869cfb000 r-xp 00004000 fd:01 2624107 /usr/lib/x86_64-linux-gnu/libapt-inst.so.2.0.0
 7f0869cfb000-7f0869cfd000 r--p 0000b000 fd:01 2624107 /usr/lib/x86_64-linux-gnu/libapt-inst.so.2.0.0
 7f0869cfd000-7f0869cfe000 ---p 0000d000 fd:01 2624107 /usr/lib/x86_64-linux-gnu/libapt-inst.so.2.0.0
...

Revision history for this message
Hans Peter (green-world78) wrote :
Revision history for this message
Christian Ehrhardt  (paelzer) wrote :

That crash seems to be from sessioninstaller which is not related to virt-manager.
But never the less from your log:
"ValueError: Namespace GtK not available"

But virt-manager could require the same.

In fact things are provided by gir1.2-gtk-3.0 and virt-manager has:
  Depends: ... gir1.2-gtk-3.0 (>= 3.10)

You could clear your /var/crash directory (IIRC by default it won't collect crashes over and over until cleared or reported). And then run virt-mamager again.
Is a new crash created, and if so how does that one look then?

Revision history for this message
Hans Peter (green-world78) wrote :
Download full text (3.2 KiB)

@ Christian, I delete all files in /var/crash and
run virt-manager again in terminal...

Segmentation fault is present.

I did this, too:

# sudo su
# PYTHONDEVMODE=1 virt-manager --no-fork

Fatal Python error: Segmentation fault

Current thread 0x00007f2309805740 (most recent call first):
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1043 in create_module
  File "<frozen importlib._bootstrap>", line 583 in module_from_spec
  File "<frozen importlib._bootstrap>", line 670 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 967 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "/usr/lib/python3.7/lzma.py", line 27 in <module>
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 728 in exec_module
  File "<frozen importlib._bootstrap>", line 677 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 967 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "/usr/lib/python3.7/shutil.py", line 29 in <module>
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 728 in exec_module
  File "<frozen importlib._bootstrap>", line 677 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 967 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "/usr/lib/python3.7/tempfile.py", line 44 in <module>
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 728 in exec_module
  File "<frozen importlib._bootstrap>", line 677 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 967 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "/usr/lib/python3/dist-packages/apport/report.py", line 12 in <module>
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 728 in exec_module
  File "<frozen importlib._bootstrap>", line 677 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 967 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "/usr/lib/python3/dist-packages/apport/__init__.py", line 5 in <module>
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 728 in exec_module
  File "<frozen importlib._bootstrap>", line 677 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 967 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "<frozen importlib._bootstrap>", line 219 in _call_with_frames_removed
  File "<frozen importlib._bootstrap>", line 953 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 983 in _find_and_load
  File "/usr/lib/python3/dist-packages/apport_python_hook.py", line 63 in apport_excepthook
Segmentation fault

I d...

Read more...

Revision history for this message
Hans Peter (green-world78) wrote :

I removed in "/usr/lib/python3/dist-packages" the file "apport_python_hook.py"

so the output of # PYTHONDEVMODE=1 virt-manager --no-fork

is now:

Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/libvirt.py", line 24, in <module>
    import cygvirtmod as libvirtmod
ModuleNotFoundError: No module named 'cygvirtmod'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/share/virt-manager/virt-manager", line 20, in <module>
    from virtinst import util
  File "/usr/share/virt-manager/virtinst/__init__.py", line 26, in <module>
    from virtinst import util
  File "/usr/share/virt-manager/virtinst/util.py", line 15, in <module>
    import libvirt
  File "/usr/lib/python3/dist-packages/libvirt.py", line 28, in <module>
    raise lib_e
  File "/usr/lib/python3/dist-packages/libvirt.py", line 20, in <module>
    import libvirtmod
ImportError: /usr/lib/x86_64-linux-gnu/libhogweed.so.4: symbol nettle_cnd_memcpy version NETTLE_6 not defined in file libnettle.so.6 with link time reference

Thomas Huth (th-huth)
no longer affects: qemu
Revision history for this message
Hans Peter (green-world78) wrote :
Revision history for this message
Christian Ehrhardt  (paelzer) wrote :

So it was stray self built libs in /usr/local - thanks for the ping here Hans.
Marking the bug as invalid then as it is no issue in the package.

Changed in virt-manager (Ubuntu):
status: Incomplete → Invalid
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.