On Wed, Oct 20, 2004 at 09:13:57AM +1000, Herbert Xu wrote:
> Phil Oester <kernel@linuxace.com> wrote:
> > Just upgraded 3 heavily used squid boxes to 2.6.9 today from
> > 2.6.8.1, and all started panicing with the same error (usually
> > within an hour of uptime):
> >
> > KERNEL: assertion (!skb_queue_empty(&sk->sk_write_queue)) \
> > failed at net/ipv4/tcp_timer.c (322)
>
> Please aplly this patch and tell us what it says just before the panic.
Ok, applied the patch, but this time didn't get the assertion error. But
managed to capture the panic on console, which may be more interesting:
Unable to handle kernel NULL pointer dereference at virtual address 00000048
printing eip:
c029f2d1
*pde = 00000000
Oops: 0000 [#1]
SMP
CPU: 0
EIP: 0060:[<c029f2d1>] Not tainted VLI
EFLAGS: 00010246 (2.6.9)
EIP is at tcp_retransmit_skb+0x41/0x320
eax: 00000000 ebx: 00000000 ecx: f6433b60 edx: 00000000
esi: f657c5a0 edi: fffffff5 ebp: f657c790 esp: f6d7be28
ds: 007b es: 007b ss: 0068
Process squid (pid: 486, threadinfo=f6d7a000 task=f6d44730)
Stack: c02971a6 00000000 0000059c f657c5a0 f657c790 f657c604 c0307ba0 c02a13cd
c0306340 c0307ba0 c03046d7 00000145 c02a4eaa c038bda0 f7185050 00000010
f6d7be8c f657c5a0 f657c790 f657c5c0 f6d7be98 c02a17c9 c201ed00 f657c5a0
Call Trace:
[<c02971a6>] tcp_enter_loss+0x76/0x240
[<c02a13cd>] tcp_retransmit_timer+0xfd/0x430
[<c02a4eaa>] tcp_v4_do_rcv+0x7a/0x110
[<c02a17c9>] tcp_write_timer+0xc9/0x100
[<c02a1700>] tcp_write_timer+0x0/0x100
[<c011ee49>] run_timer_softirq+0xd9/0x170
[<c011b0f6>] __do_softirq+0xb6/0xd0
[<c011b13d>] do_softirq+0x2d/0x30
[<c010de35>] smp_apic_timer_interrupt+0x85/0xf0
[<c010492e>] apic_timer_interrupt+0x1a/0x20
[<c029147c>] tcp_poll+0xcc/0x170
[<c0147cb2>] fget+0x42/0x60
[<c0272122>] sock_poll+0x12/0x20
[<c015905d>] do_pollfd+0x8d/0xa0
[<c01590d2>] do_poll+0x62/0xc0
[<c015925f>] sys_poll+0x12f/0x210
[<c01069b5>] do_IRQ+0x105/0x120
[<c0158620>] __pollwait+0x0/0xd0
[<c0103eed>] sysenter_past_esp+0x52/0x71
Code: 31 d2 e8 23 f6 ff ff 89 44 24 08 8b 46 78 89 c2 c1 fa 02 01 d0 8b 96 84 00 00 00 39 c2 0f 4f d0 8b 46 60 39 d0 0f 8f b9 01 00 00 <8b> 53 48 8b 4d 10 39 ca 79 55 39 4b 4c 79 08 0f 0b 05 04 90 46
<0>Kernel panic - not syncing: Fatal exception in interrupt
Phil
-
: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
