On Wed, Oct 28, 2009 at 1:05 PM, Brooks Lyrette
<brooks.lyrette@gmail.com> wrote:
> The machine is running a moderate load. This is running on a Solaris Zone.
>
> Memory: 32G phys mem, 942M free mem, 76G swap, 74G free swap
>
> PID USERNAME LWP PRI NICE SIZE RES STATE TIME CPU COMMAND
> 5069 postgres 1 52 0 167M 20M sleep 0:04 13.50% postgres
Hm, well 400 processes if each were taking 190M would be 76G. But that
doesn't really make much sense since most of the 167M of that process
is presumably the shared buffers. What is your shared buffers set to
btw? And your work_mem and maintenance_work_mem?
Fwiw ENOMEM is documented as "There is not enough swap space.".
Perhaps you have some big usage spike which uses up lots of swap and
causes postgres to start needing lots of new processes at the same
time?
--
greg
