Yesterday my linode was not responding, and the control panel said if was switched off. I booted thru the control panel and all was well.

Now it happened again today 4 times. It ran for a few minutes before being powered off. I downgraded from the 2.6 to 2.4 kernel and its been working OK for 10 hours.

The linode was host2/Redhat large pofile/2.6 kernel. The log messaged showed nothing when it stopped. Setup had been changed in months.

I've seen this happen very occasionally in the past but 4 times in 2 hours is too much. Has anyone else experienced this?

"logview" via Lish shows the following panic:

Kernel panic - not syncing: Kernel mode fault at addr 0x8, ip 0x400fa148
 
EIP: 0073:[<400fa148>] CPU: 0 Not tainted ESP: 007b:bffffd88 EFLAGS: 00000246
    Not tainted
EAX: 00000001 EBX: 00000000 ECX: bffffdab EDX: 00000001
ESI: 0804ad00 EDI: bffffdab EBP: bffffdb8 DS: 007b ES: 007b
Call Trace: 
 [<80034549>] vprintk+0xe9/0x13c
 [<800421ee>] notifier_call_chain+0x1e/0x38
 [<80033a9e>] panic+0x56/0xbc
 [<8001d907>] segv+0x83/0x1c8
 [<8001d9d6>] segv+0x152/0x1c8
 [<8030f1d7>] sigemptyset+0x17/0x30
 [<8001bd91>] change_signals+0x41/0x6c
 [<8001dd3c>] segv_handler+0x160/0x208
 [<8001dcab>] segv_handler+0xcf/0x208
 [<80022d2c>] sig_handler_common_skas+0x7c/0x98
 [<80022d45>] sig_handler_common_skas+0x95/0x98
 [<8001de2a>] sig_handler+0x32/0x34
 [<8030edfd>] __libc_longjmp+0x3d/0x50
 [<8030eded>] __libc_longjmp+0x2d/0x50
 [<80022eda>] do_buffer_op+0x76/0x120
 [<8030eef8>] __restore+0x0/0x8
 [<80023963>] chan_window_size+0x7/0x44
 [<8001ebc2>] setjmp_wrapper+0x16/0x50
 [<8030f1d7>] sigemptyset+0x17/0x30
 [<8001bed4>] set_signals+0x6c/0x10c
 [<8030f1d7>] sigemptyset+0x17/0x30
 [<8001bd91>] change_signals+0x41/0x6c
 [<8030f1d7>] sigemptyset+0x17/0x30
 [<8030f1d7>] sigemptyset+0x17/0x30
 [<8001bed4>] set_signals+0x6c/0x10c
 [<801fee85>] tty_read+0xa5/0xf0
 [<802043fc>] write_chan+0x0/0x29c
 [<80065b5b>] vfs_read+0xc3/0xf4
 [<80065d97>] sys_read+0x3b/0x68
 [<800228a6>] execute_syscall_skas+0xaa/0xec
 [<80025149>] winch_interrupt+0x49/0xa8
 [<80017047>] handle_IRQ_event+0x27/0x6c
 [<800171e0>] do_IRQ+0x68/0xc4
 [<80022d84>] user_signal+0x3c/0x44
 [<80021efb>] userspace+0x18f/0x1e4
 [<8030eef8>] __restore+0x0/0x8
 [<8030f151>] kill+0x11/0x20

I'll report this to the uml-devel mailing list, however judging by the chan_window_size and winch_interrupt system calls, I believe this is the bug that Newsome has been hunting for months, and has been recently fixed in UML. I'll try to get a new 2.6-um kernel out in the next week or two.

-Chris