Fatal trap 12: page fault panic with recent kernel with ZFS

[Adam McDougall]

On Mon, May 18, 2009 at 07:03:45PM -0700, Kip Macy wrote:

  > Thanks, I appreciate all this work. ?Not allowing inactive pages to
  > shrink the ARC sounds great as an option. ?I would be willing to bet
  > that allowing inactive pages to shrink the arc would be far less
  > detrimental to most people who aren't running a constant busy file
  > server load, and its definitely important to try to protect untuned
  > boxes.
  
  Allowing NFS to use ARC buffers might be one solution to that.

That would be interesting.  I haven't used ZFS for any NFS serving
yet though.  With my mix of userspace file serving daemons my Inactive
mem just rises 1-3M/sec until almost all free memory is consumed and
I don't know why.  None of the processes in top or ps can account for
16G of Inactive.    
  
  > Do you have any suggestions for increasing the amount of memory ARC
  > can use? ?I've had difficulty increasing kmem past a few gigs on any
  > of my recent builds (all past where kmem was changed so it could be
  > more than ~2g) because at some point the kernel would stop booting.
  > If I increase them too far, a few lines of the booting kernel would
  > print, followed by a long stream of page fault panics or something
  > with a sudden reboot. ?With the recent change allowing the use of
  > direct mem, the ARC could easily use ample memory except it turned
  > out not to be stable.
  
  As of r192216 that should not be a probably any more. The maximum kmem
  is now 512GB.  It will be at least a year or two before anyone bumps
  his head against that.
  
Ahh oops, I mistakenly thought it was backed out a few minutes ago
but that was something else.  I guess that gives me something else I
can test.  A number of changes went in recently and it was hard to tell
which commits were causing which symptoms.