System crashes if disconnected from network
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux-source-2.6.20 (Ubuntu) |
Won't Fix
|
Medium
|
Unassigned |
Bug Description
After upgrading to Feisty, I have an interesting new feature where if the box is disconnected from the Network then the computer freezes completely (music stops playing, won't respond to mouse input, won't respond to keyboard input, gui freezes, etc.) and the caps lock and scroll lock leds flash. This has occurred several times but is not reproducible everytime. It has only started to occur after I upgraded to Feisty.
Whats even more scary is that there have also beens time when this has occurred but the above mentioned trigger did _not_ occur! ie. Something else caused it to freeze up like this.
I'd be happy to help test and debug this anyway I can. I tried looking in the log files for some indication of an error but none were found.
description: | updated |
description: | updated |
Changed in linux-source-2.6.20: | |
assignee: | nobody → ubuntu-kernel-team |
importance: | Undecided → Medium |
Changed in linux-source-2.6.20: | |
assignee: | brian-murray → nobody |
status: | Incomplete → Won't Fix |
This evening I was fooling around with this to see if I could uncover any more clues (ie. I had noticed that music was usually playing when the crash occurred so I tried unplugging it with different players running in attempt to get it to crash again - but this was unsuccessful) . I found out that it does not occur every time the network is disconnected (description updated).
At the end of the night, I tried unplugging it and plugging it back one more time (nothing happened) before I clicked Applications > Quit in Xubuntu and clicked shutdown. After X closed, I was presented with scrolling text that repeated the following:
"atkdbd.c: Spurious ACK on isa0060/serio0. Some program might be trying to access hardware directly."
At the same time, the caps and scroll lock leds were flashing (like in the other freezes). A fellow on IRC had suggested that I try a few different key combinations with alt+ctrl+ shift+sysrq+ <blah> or something to that effect and I tried it with b (which he told me, and which I lated tried for myself to success, rebooted the system). At this time, I was presented with something that looked like a traceback at the top of the screen and two forward slashes at the bottom of he screen. The leds had stopped blinking and no input was the keyboard seemed to register.
The traceback was as following (though I copied it by hand so there might be some user errors):
[<d0c4c060>] device_cmp+0x0/0x50 [ipt_MASQUERADE] event+0x1d/ 0x20 [ipt_MASQUERADE] call_chain+ 0x25/0x30 notifier_ call_chain+ 0x25/40 del_ifa+ 0x11f/0x220 ioctl+0x28c/ 0x6c0 0xbf/0x210 0x0/0x210 0x5c/0x2a0 0x5c/0x2a0 past_esp+ 0x5f/0x89 0x1b3/0x580
=== ======= ======= ======= ===
[<d0c5c04d>] musq_inet_
[<c02e8655>] notifier_
[<c0131ab5>] blocking_
[<c02bfa4f>] __inet_
[<c02c0c6e>] devinet_
[<x026e25f >] sock_ioctl+
[<c026ela0 >] sock_ioctl+
[<c017eb0b>] do_ioctl+0x2b/0x90
[<c017ebcc>] vfs_ioctl+
[<c017ee8c>] sys_ioctl+
[<c0103116>] sysenter_
[<c02e0033>] vcc_connect+
I hope this helps and sorry if I cause you any due confusion and stress is I made a typo in there! If the above incident is a separate bug, please let me know and I'll create a new bug report.