On 9 February 2010 00:37, Ksenia <[email protected]> wrote:
>
>
> On Feb 7, 11:37 pm, Graham Dumpleton <[email protected]>
> wrote:
>> On 6 February 2010 14:50, Jason Garber <[email protected]> wrote:
>>
>>
>>
>>
>>
>> > Hi Ksenia,
>> > Yes, this is possible. What you need to do is run a single DaemonProcess
>> > named, say, "Primary". Then within each VirtualHost, you need
>> > WSGIProcessGroup that refers to "Primary".
>> > Something like this:
>> > WSGIApplicationGroup %{GLOBAL}
>> > WSGIDaemonProcess Primary threads=15 python-path=/foo/bar/Python
>> > <VirtualHost *:80>
>> > ServerName foo1.com
>> > WSGIProcessGroup Primary
>> > WSGIScriptAlias /User /foo/bar/app.wsgi
>> > </VirtualHost>
>> > <VirtualHost *:80>
>> > ServerName foo2.com
>> > WSGIProcessGroup Primary
>> > WSGIScriptAlias /User /foo/bar/app.wsgi
>> > </VirtualHost>
>>
>> You can actually put the WSGIProcessGroup outside of the VirtualHost
>> directive.
>>
>> Thus:
>>
>> WSGIDaemonProcess Primary threads=15 python-path=/foo/bar/Python
>> WSGIProcessGroup Primary
>> WSGIApplicationGroup %{GLOBAL}
>>
>> <VirtualHost *:80>
>> ServerName foo1.com
>> WSGIScriptAlias / /foo/bar/app.wsgi
>> </VirtualHost>
>> <VirtualHost *:80>
>> ServerName foo2.com
>> WSGIScriptAlias / /foo/bar/app.wsgi
>> </VirtualHost>
>>
>> I used '/' with WSGIScriptAlias as you asked for case where mounted at
>> root of site.
>>
>> BTW, if the whole VirtualHost goes through to this application and
>> nothing else is served by them, and every virtual host on server
>> follows this same recipe, you could do away with the VirtualHost
>> directives altogether and just put WSGIScriptAlias at global scope.
>
>
> Thank you for answering. In our situation not all VirtualHost that are
> hosted on the server are using the same application. So I guess
> WSGIScriptAlias should go inside every VIrtualHost.
> On another server we are using mod_wsgi for a different setup (one
> virtualhost, one app), with the startup script that looks like:
>
> #!/usr/bin/env python
> import os
> from paste.deploy import loadapp
> from pkg_resources import resource_filename
> app_path = os.path.dirname(resource_filename('myapp', ''))
> if '/test/' in app_path:
> config_file = 'test.ini'
> else:
> config_file = 'live.ini'
> config_path = os.path.join(app_path, config_file)
> application = loadapp('config:' + config_path)
>
>
> But I think if I use this script in the situation with 300
> VirtualHost, the same application will be loaded 300 times?! Is it how
> it supposed to work?
So long as you have outside of VirtualHost:
WSGIDaemonProcess Primary threads=15 python-path=/foo/bar/Python
to define the daemon process group. And then:
WSGIScriptAlias / /some/path/app.wsgi
WSGIProcessGroup Primary
WSGIApplicationGroup %{GLOBAL}
inside all VirtualHost's which need it, since not all should be
delegated you are now saying, then there will be only once instance of
application loaded.
This is because WSGIProcessGroup says handle any WSGI applications in
this context in that named process group.
The WSGIApplicationGroup says for the target process, handle any WSGI
applications in this context in the same Python interpreter within
that process.
Thus, so long as all WSGIScriptAlias's refer to the exact same script
file, every virtual host will use that same file and it will only be
loaded once.
Graham
--
You received this message because you are subscribed to the Google Groups
"modwsgi" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/modwsgi?hl=en.
