Re: RFC: nexthop cache separation

[Ryota Ozaki <ozaki-r%netbsd.org@localhost>]

On Wed, Apr 6, 2016 at 9:40 PM, Christos Zoulas <christos%astron.com@localhost> wrote:
> In article <CAKrYomjsmwB_Ogj8gPyf_Bxd8EMgyrAAkWn+TFqGGKC+Q5DHXw%mail.gmail.com@localhost>,
> Ryota Ozaki  <ozaki-r%netbsd.org@localhost> wrote:
>>On Wed, Apr 6, 2016 at 1:57 AM, Christos Zoulas <christos%astron.com@localhost> wrote:
>>> Since these changes, my machine crashes doing ufs lookups in /var/run
>>> for any daemons... This happens only if I turn nfs_client=YES in rc.conf.
>>>
>>> Seems something is corrupting memory related to nfs.
>>
>>Hmm, it works for me (with a latest kernel and old userland binaries).
>>Can you provide dmesg and backtrace?
>
> http://www.zoulas.com/NetBSD/var.jpg
> http://www.zoulas.com/NetBSD/dmesg

Heh, the backtrace is completely in fs...

>
> Is yours an nfs server?
> Are you using nfs clients?

I set up an nfs client on NetBSD and an nfs server was Linux
(that I can set up quickly :-/).

So you set up an nfs server and nfs client(s) and
the panic happens on the server, right?

  ozaki-r