On Wed, 29 Sep 2010 11:57:10 +0200, you wrote: >sigh. use-after-free (most likely, at least) somewhere. unlikely to be >carp itself. might be re (wild guess).
I think your "somewhere near re" theory might have some merrit to it. :) I've had a number of crashes over the past couple days, but this one stood out and didn't look like the others. login: uvm_fault(0xd0a05960, 0x0, 0, 1) > e kernel: page fault trap, code=0 Stopped at ether_input+0xa9: testb $0x1,0(%edx) ddb> trace ether_input(d1be7054,0,d66ca000,0,60) at ether_input+0xa9 re_rxeof(d1be7000,dbc13000,3e,4001,1) at re_rxeof+0x33c re_intr(dbe7000) at re_intr+0x12a Xintr_ioapic2() as Xintr_ioapic2+0x62 --- interrupt --- cpu_idle_cycle(d0ac1aa0) at cpu_idle_cycle+0xf Bad frame pointer: 0xd0b75e48 Full crash with ps at: http://webpages.charter.net/cycler/crash4.jpg
