On Mon, 13 Jun 2005 14:30:22 +0800, Uwe Dippel wrote: > ip_output(d21e6000,0,d05b90a0,1,0,0,0,0) at ip_output+0x908 > ip_forward(d21e6000,1,0,50,d07b204c) at ip_forward+0x188 > ipv4_input(d21e6000,d07b4ec0,0,d06d2000) at ipv4_input+0x25a > ipintr(58,10,10,10,d06d2000) at ipintr+0x67 > Bad frame pointer: 0xd06d3e74
Looks rather similar to another problem, when the kernel trips as well from communication with an interface: http://www.monkey.org/openbsd/archive2/bugs/200505/msg00031.html There it goes like this: ip_output(d4de4600,0,d05b90a0,1,0,0,0,d4e57000) at ip_output+0x908 ip_forward(d4de4600,1,0,50,d4de4600) at ip_forward+0x188 ipv4_input(d4de4600,d0a3c900,0,d06d2000) at ipv4_input+0x25a ipintr(58,10,10,10,d06d2000) at ipintr+0x67 Bad frame pointer: 0xd06d3e74 Seems to be a real problem affecting more than one piece of hardware and interface. (ural versus rl) In both cases it happens on old, low-cost boxes. Similar to that problem, also here the crashes are somewhat random; only that it doesn't take minutes but seconds; only a few packets are usually sent here before the fault sets in. I hope someone with much more insight looks into this, Uwe
