Andreas Jung wrote:


--On Mittwoch, 13. April 2005 10:51 Uhr +0200 Lucas Hofman <[EMAIL PROTECTED]> wrote:

I am running a server with a handfull of plone intranet sites. The server
has 4GB  memory, but is now starting to page.

What I notice is that the different Plone instances vary widely in there
memory usage. For one site every process uses 400Mb, and there are 6
processes running! That is 2.4 Gbyte for one site!


Zope runs only in one process with multiple threads (where threads on some OSes
are implemented as processes :-))


I was tricked by the output of ps and top to believe that all these were processes. In addition, looking at the memory usage with 'free' gave me more insight in what linux is actually doing. Thanks.



Restarting Zope regularly (at night) might be a wise choice (this is
a *serious* advice).


Will do. A bit of a nuisance with offices and vessels all over the globe accessing the site. Actually the 400GB process was only running for one day. Was there not a Zope version with some serious memory leaks that were fixed? (these sites are running different python/zope/plone versions)


Lucas

_______________________________________________
Zope maillist - Zope@zope.org
http://mail.zope.org/mailman/listinfo/zope
** No cross posts or HTML encoding! **
(Related lists - http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope-dev )

Reply via email to