> as we tested scaling and put more load on the servers (i.e. allowing
> Apache to spawn more children), we were surprised to see that the system
> doesn't tend to go below 8Gb (half) of the available RAM. Further requests
> for memory allocation failed (can't fork new processes including new sshd,
> vmstat, top and bash).
Jim, I'm not an expert by any means, but first of all, I recommend you use the
"prstat -s size" command instead of the "top" command to see what's going on in
the system (prstat is more accurate and much more scalable). Also, when this
weird swap thing that you are talking about happens, can you use the "su"
command to assume root priveleges and run this command (it will take a long
time for this command to finish running):
echo "::memstat" | mdb -k
and post the results in this thread? Hopefully it will contain some useful
information about what is actually going on inside your system.
Also, here are some links about ZFS RAM usage that might point you in the right
direction about where the kernel tunable parameters are:
http://blogs.sun.com/lalith/entry/why_opensolaris_is_a_glutton
http://solarisdesktop.blogspot.com/2008/11/where-is-memory-gone.html
--
This message posted from opensolaris.org
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
