I didn't know if this should go here or Abends/Hangs, but the server's
not abended...yet. This is a Dell PowerEdge 2850, dual-P3/1.0 Ghz, 4GB
RAM, four 72-GB SCSI drives in Raid5+1 configuration. This box is our
primary GroupWise server as well, and runs the MTA, GWIA, WebAccess,
and two POAs.

Typically (I check daily as part of maintenance) the cache buffers on
this box run around 45-46%. We reboot servers once a month and then
the cache buffers go back up to 55%, then creep back down over the next
week or so and settle at 45% or so. Yesterday the cache buffers were
at 55% (we rebooted the servers on Friday).

Today I'm running maintenance and the cache buffers are at 28%! Before
I optimized our servers they would sometimes run around 35% but since
then they've been running over 50%, but I have NEVER seen one of the
servers (let alone the one with the most RAM) BELOW 30%. I checked the
Allocated memory pool and over 200MB is allocated to NSS.NLM (I'm not
sure what it should be but that seems excessive). Anything I should
check before I decide to restart this pig? Given that over 500 people
use it for email, I probably won't have a chance to IPL it until after
5 PM tonight, provided it doesn't crash first.

Thanks in advance.