Sarton O'Brien wrote:
Joel CARNAT wrote:Looks like it may be the same problem I have encountered (see thread 'dom0 panic on domu dettach'). If I shutdown the domu, the vnd slot was still occupied and forcing an unmount would panic the dom0. Rebooting the domu would cause the panic straight off.#################### uvm_fault(0xd0d67dd0, 0, 1) -> 0xe kernel: supervisor trap page fault, code=0Stopped in pid 26804.1 (ls) at netbsd: turnstile_block+0x411: movl 0x10(%eax), %eax#################### My dom0 is 4.99.49 GENERIC (plus wpi* enabled). Attached is the domU conifguration file.I don't use hvm so I could be way off.I was going to try and obtain more information but have just updated to the latest kernel as of yesterday. I'll report back in a minutes.
OK, all looks to be good again.NetBSD gogeta.internal 4.99.52 NetBSD 4.99.52 (XEN3_DOM0) #0: Mon Feb 4 18:35:23 EST 2008 root%spike.internal@localhost:/usr/obj/sys/arch/i386/compile/XEN3_DOM0 i386
I imagine you'd have needed to be utilising a vnd backend to experience this problem, which may account for so few people experiencing it.
I can't vouch for the state of the kernel since last night when I updated but if you'd like the one I'm using available somewhere, let me know.
Sarton