SUMMARY: How big can max_per_proc_address_space/data_size actually be?

From: Reed, Judith (jreed@navisite.com)
Date: Thu Oct 23 2003 - 14:34:42 EDT

• Next message: Brehl, Blake: "korn shell script question"
• Previous message: Fred Schenerman: "Problem in Upgrading from version 5.1a to 5.1b"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

I asked about guidelines for:
[max_]per_proc_address_space
and
[max_]per_proc_data_size

Thanks to all who replied.
Jeffrey Hummel reported his extremely high identical values under 4.0g:
          max-per-proc-address-space = 68719476736
        max-per-proc-data-size = 68719476736
        per-proc-address-space = 68719476736
        per-proc-data-size = 68719476736

Alan Rollow sent a long and interesting discussion of things to
consider, that I can pass along to anyone who needs it.

John Lanier sent the commands to get the max values:
        sysconfig -Q proc max_per_proc_address_space
        sysconfig -Q proc max_per_proc_data_size
and pointed me to "man sys_attrs_proc", which gives default, min and max
values for all 4 params.

In the end, for this test system, I went with 10GB for the max_* values,
and 8GB for the per_proc* values. I also turned off eager swap. This is
on an 8GB memory system. We'll see how that goes.

Thanks again!!!

Judith Reed
jreed@navisite.com
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
Benjamin Franklin

• Next message: Brehl, Blake: "korn shell script question"
• Previous message: Fred Schenerman: "Problem in Upgrading from version 5.1a to 5.1b"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:39 EDT