vm_thread_new: kstack allocation failed with vm.kmem_size="1536M"

[Pete French]

So, I have a farm of machines runnign 7.1/amd64, all of which have 16 gig of
memory in them. This afternoon, as an experiment, I altered loader.conf
to have these two lines in it:

vm.kmem_size="1536M"
vm.kmem_size_max="1536M"

This is what I do on machines running ZFS - these machines are not, however
running ZFS, and do not have the zfs module loaded. I just wanted to see if
they would run OK with those kernel settings (as I may put ZFS on them in
the future)

I expected it to run fine, I just wanted to make sure. But after about
an hour I started getting the message in the subject line, and the
machines were unable to fork and needed to be reset.

Explanation anyone ? This makes no sense to me - I have actually expanded
the amount of memory available, so why is it now running out of stack
space ?! The machines are running a very simple setup of apache, mysql
and tomcat - and this runs fine with the kernel variables set as default.

I am very puzzled.

-pete.