Re: share memory fragmentation
From: joel garry <joel-garry_at_home.com>
Date: Mon, 22 Jun 2009 10:06:54 -0700 (PDT)
Message-ID: <19d626e0-b7bb-4b59-b748-66b4a324cb36_at_x1g2000prh.googlegroups.com>
On Jun 21, 12:03 am, patrick boulay <boul..._at_free.fr> wrote:
> has anyone came accross a share memory fragmentation ?
> We have this issue sometime with a 300G SGA instance ( non RAC )
Date: Mon, 22 Jun 2009 10:06:54 -0700 (PDT)
Message-ID: <19d626e0-b7bb-4b59-b748-66b4a324cb36_at_x1g2000prh.googlegroups.com>
On Jun 21, 12:03 am, patrick boulay <boul..._at_free.fr> wrote:
> has anyone came accross a share memory fragmentation ?
> We have this issue sometime with a 300G SGA instance ( non RAC )
See this about what to post: http://dbaoracle.net/readme-cdos.htm#subj12
In addition to what Mark mentioned, note that different platforms and configurations handle large SGA's quite differently. Sometimes a single line like turning NUMA on or off or keeping an undocumented parameter from an earlier patch set can be enough to make things work totally differently. So, have you changed anything? Has it ever worked?
jg
-- _at_home.com is bogus. http://joinlapd.com/images/left_nav_image.gifReceived on Mon Jun 22 2009 - 12:06:54 CDT
