Investigating a system "freeze"
Am on
- Ubuntu 10.04.2 LTS
- Linux skiran 2.6.32-32-generic-pae #62-Ubuntu SMP Wed Apr 20 22:10:33 UTC 2011 i686 GNU/Linux
My machine froze for the 5th time in 2 days.
- Had to reboot, and login to a failsafe session to collect and review logs.
- ALT+F2 to run nm-applet, so that the network connectivity works
- ALT+F2 to run gnome-do, to launch my applications (I prefer using gnome-do, to using menu options, or ALT+F2)
I noticed the following errors in the various /var/log/* files, at the time my machine last froze (around 14:32 on June 10)
In kern.log
Jun 10 14:32:17 skiran kernel: [ 0.766164] PM: Resume from disk failed.
In syslog
Jun 10 14:32:17 skiran kernel: [ 0.766164] PM: Resume from disk failed.
Jun 10 14:32:22 skiran pulseaudio[1867]: module.c: Failed to load module "module-alsa-card" (argument: "device_id="1" name="pci-0000_01_00.1" card_name="alsa_card.pci-0000_01_00.1" tsched=yes ignore_dB=no card_properties="module-udev-detect.discovered=1""): initialization failed.
Jun 10 14:32:22 skiran pulseaudio[1867]: module.c: Failed to load module "module-alsa-card" (argument: "device_id="29" name="platform-thinkpad_acpi" card_name="alsa_card.platform-thinkpad_acpi" tsched=yes ignore_dB=no card_properties="module-udev-detect.discovered=1""): initialization failed.
Jun 10 14:32:24 skiran gdm-session-worker[1865]: GLib-GObject-CRITICAL: g_value_get_boolean: assertion `G_VALUE_HOLDS_BOOLEAN (value)' failed
Jun 10 14:32:35 skiran pulseaudio[2062]: module.c: Failed to load module "module-alsa-card" (argument: "device_id="1" name="pci-0000_01_00.1" card_name="alsa_card.pci-0000_01_00.1" tsched=yes ignore_dB=no card_properties="module-udev-detect.discovered=1""): initialization failed.
Jun 10 14:32:35 skiran pulseaudio[2062]: module.c: Failed to load module "module-alsa-card" (argument: "device_id="29" name="platform-thinkpad_acpi" card_name="alsa_card.platform-thinkpad_acpi" tsched=yes ignore_dB=no card_properties="module-udev-detect.discovered=1""): initialization failed.
Also in wcstatus.log noticed several errors, occurring at the time the machine froze. This is the IBM VPN client.
3053452144 : device : (2011-06-10 / 14:32:20.385908) : wg_status_linux_hw_name: wlan0 is a PCI interface.
3053452144 : device : (2011-06-10 / 14:32:35.043054) : wg_status_adapter_event_thread: broke out of read for IP interface state changes.
3053452144 : device : (2011-06-10 / 14:32:35.043086) : wg_status_adapter_event_thread: calling event function so the event can be processed.
3053452144 : device : (2011-06-10 / 14:32:35.043124) : wg_status_gather_unix_add_new_ifaces: see if we have new ifaces
3053452144 : device : (2011-06-10 / 14:32:35.043263) : wg_status_gather_linux_if_info: The ifr struct from hwaddr ioctl for adapter eth0 is
0x0000 : 65 74 68 30 00 00 00 00 00 00 00 00 00 00 00 00 [eth0............]
0x0010 : 01 00 f0 de f1 0d 40 d2 c8 d9 ff b5 b1 b0 83 b7 [......@.........]
.
3053452144 : device : (2011-06-10 / 14:32:35.043293) : wg_status_gather_linux_if_info: The adapter description for adapter eth0 is Ethernet.
3053452144 : device : (2011-06-10 / 14:32:35.043432) : wg_status_linux_hw_name: eth0 is a PCI interface.
3053452144 : device : (2011-06-10 / 14:32:35.046584) :
<Name: eth0 >
<Hardware Address: f0:de:f1:0d:40:d2 ><IP Address: 9.182.74.162 ><Subnet: 255.255.255.0 >
<Network MTU: 1500 ><State: active >
<Adapter Description: Intel Corporation 82577LM Gigabit Network Connection >
<Adapter Speed: 100000000 ><POINT-TO-POINT: false >
3053452144 : device : (2011-06-10 / 14:32:35.046611) : <Name: eth0 >
<Received Stats: (1471 Byte)(0009 Pack)(0000 Err) >
<Transmit Stats: (1836 Byte)(0010 Pack)(0000 Err) >
3053452144 : device : (2011-06-10 / 14:32:35.046631) : wg_status_gather_impl::wg_consumer_add_item added object to consumer signal the consumer thread.
3053452144 : device : (2011-06-10 / 14:32:35.046651) : wg_status_gather_linux_if_info: The ifr struct from hwaddr ioctl for adapter wlan0 is
0x0000 : 77 6c 61 6e 30 00 00 00 00 00 00 00 00 00 00 00 [wlan0...........]
0x0010 : 01 00 00 27 10 7c 6f e8 00 00 00 00 5e f7 88 b7 [...'.|o.....^...]
.
3053452144 : device : (2011-06-10 / 14:32:35.046674) : wg_status_gather_linux_if_info: The adapter description for adapter wlan0 is Ethernet.
3053452144 : device : (2011-06-10 / 14:32:35.046783) : wg_status_linux_hw_name: wlan0 is a PCI interface.
I have attached the kern.log, syslog, wcstatus.log, messages, dpkg.log (on pastebin.ubuntu.com. Contains messages generated on 10 June only) - thanks for the tip Jamess
Updates: 12 June I logged into gnome session, and had enabled Apport. Two minutes after logging in, the system froze. Logged back into fail-safe, and noticed these two crash reports in /var/crash: _usr_bin_skype.1000.crash, _usr_lib_gnome-do_Do.exe.1000.crash. Also /var/log/apport.log reported process 2222 caused signal 6. This process happens to be mono+gnome-do combination.
Am thinking this could be some problem related to mono - but then am not 100% sure. Referred to this link, and ran this command sudo apt-get purge libmono0 mono-common libgdiplus
. Need to see if this makes a difference.
Please let me know if you find anything amiss, and can suggest a solution. I do not want to leave the failsafe session, till I come across some possible solutions.
I noticed that my system freeze happened every time I had Firefox open - totally weird, I know. FF was at 4.x level. But then, I was desperate to get a resolution for the system freeze problem; couldn't possibly work in the gnome-safe mode forever. I noticed that my update-manager did not complete installation of a set of updates, and that it always stopped at the "ubufox" update. ubufox - Ubuntu Firefox specific configuration defaults and apt support
Went into synaptic, removed ubufox, and noticed that Firefox then updated itself to 5.0. Also, a whole load of other updates happened: http://pastebin.ubuntu.com/640006/.
So, as it stands now, my machine has been functioning ok, in the regular gnome-session. Still not sure what was causing the problem, because I didn't come across some conclusive evidence. Closing this thread, by marking this as the answer.