On Tuesday, March 20, 2012 10:28:49 AM UTC, Another Django Newbie wrote:
>
>
>
> On Tuesday, March 20, 2012 10:01:36 AM UTC, Tom Evans wrote:
>>
>> On Mon, Mar 19, 2012 at 5:24 PM, Jani Tiainen <xxxxxxxxxxxxxx> wrote:
>> > Hi,
>> >
>> > Since we use same setup except one part: we use mod_wsgi instead of
>> > mod_fcgi.
>> >
>> > (Since wsgi is considered to be defacto protocol). Could you try to use
>> it?
>> >
>>
>> WSGI is the de-facto for hosting python in web servers. If you aren't
>> running just python, it's nice to use the same hosting mechanism for
>> all your apps, and fastcgi is a supported mechanism.
>>
>> OP: You mention using mod_fcgid, but you do not give us any idea of
>> how you are using mod_fcgid. Configuration and error logs please. FYI,
>> we run django quite happily under mod_fastcgi.
>>
>> Cheers
>>
>> Tom
>>
>
>
> Hi Tom, Jani,
>
> Currently struggling to compile mod_wsgi, or more specifically a shared
> library version of python - something to do with mismatched 32/64 bit
> libraries I think.
>
> Here's an extract from httpd.conf.
>
> RewriteEngine On
> RewriteRule ^.*/cfServer/(.*)$ /cgi-bin/django.cgi/$1 [PT]
> <Directory "/usr/local/apache2/cgi-bin">
> Options ExecCGI
> AddHandler fcgid-script .cgi
> # AddHandler cgi-script .cgi
> Require all granted
> </Directory>
>
> and the relevant django.cgi
>
> #!/usr/local/bin/python
> from FcgiWsgiAdapter import list_environment, serve_wsgi
>
> import os, sys
> os.environ['DJANGO_SETTINGS_MODULE'] = 'cfServer.settings'
> os.environ['ORACLE_HOME'] = '/oracle/11gStdEd'
>
> from django.core.handlers.wsgi import WSGIHandler
>
> # use this to test that fastcgi works and to inspect the environment
> # serve_wsgi(list_environment)
> # use this to serve django
> serve_wsgi(WSGIHandler())
>
> The "FcgiWsgiAdapter" referred to came from
> djangosnippets.org<http://djangosnippets.org/snippets/1307/>
> .
>
> As to error logs, as I mentioned in the original post I don't see anything
> added to any logs when this happens and there is only a warning about the
> ssl library when apache is restarted.
>
> Regards, ADN
>
*So I swapped to using WSGI with this apache config ....*
WSGIScriptAlias / /home/user/python/djcode/cfServer/apache/django.wsgi
WSGIDaemonProcess site-1 threads=30
WSGIProcessGroup site-1
<Directory "/home/user/python/djcode/cfServer/apache">
AllowOverride None
Options None
Require all granted
</Directory>
and this django.wsgi
import os, sys
sys.path.append('/home/user/python/djcode')
os.environ['DJANGO_SETTINGS_MODULE'] = 'cfServer.settings'
os.environ['ORACLE_HOME'] = '/oracle/11gStdEd'
os.environ['PYTHON_EGG_CACHE'] = '/var/www/.python-eggs'
import django.core.handlers.wsgi
application = django.core.handlers.wsgi.WSGIHandler()
and I get segmentation faults (debug level tracing from apache error log)
[Tue Mar 20 14:16:28.126341 2012] [core:notice] [pid 17487:tid
47401308446624] AH00051: child pid 17490 exit signal Segmentation fault
(11), possible coredump in /tmp/apache2-gdb-dump
[Tue Mar 20 14:16:28.126385 2012] [:info] [pid 17487:tid 47401308446624]
mod_wsgi (pid=17490): Process 'site-2' has died, deregister and restart it.
[Tue Mar 20 14:16:28.126394 2012] [:info] [pid 17487:tid 47401308446624]
mod_wsgi (pid=17490): Process 'site-2' has been deregistered and will no
longer be monitored.
[Tue Mar 20 14:16:28.126826 2012] [:info] [pid 17846:tid 47401308446624]
mod_wsgi (pid=17846): Starting process 'site-2' with uid=48, gid=48 and
threads=25.
[Tue Mar 20 14:16:28.127268 2012] [:info] [pid 17846:tid 47401308446624]
mod_wsgi (pid=17846): Initializing Python.
Back trace on core file ...
Core was generated by `/usr/local/apache2/bin/httpd -k start'.
Program terminated with signal 11, Segmentation fault.
#0 0x00000036422cb2e6 in poll () from /lib64/libc.so.6
(gdb) bt
#0 0x00000036422cb2e6 in poll () from /lib64/libc.so.6
#1 0x00002b1c7a05dd79 in apr_poll (aprset=0x7fff13f01e40, num=1,
nsds=0x7fff13f01e94, timeout=0) at poll/unix/poll.c:120
#2 0x00002b1c7d5166d3 in wsgi_daemon_main (p=0x1257a138, daemon=0x12696218)
at mod_wsgi.c:11330
#3 0x00002b1c7d51851f in wsgi_start_process (p=0x1257a138,
daemon=0x12696218)
at mod_wsgi.c:11969
#4 0x00002b1c7d518f50 in wsgi_start_daemons (pconf=0x1257a138,
ptemp=<value optimized out>, plog=<value optimized out>,
s=<value optimized out>) at mod_wsgi.c:12181
#5 wsgi_hook_init (pconf=0x1257a138, ptemp=<value optimized out>,
plog=<value optimized out>, s=<value optimized out>) at mod_wsgi.c:13755
#6 0x0000000000444c2c in ap_run_post_config (pconf=0x1257a138,
plog=0x125df538, ptemp=0x125a5348, s=0x125a3508) at config.c:101
#7 0x0000000000428734 in main (argc=3, argv=0x7fff13f02498) at main.c:765
Using WSGI briefly solved the original problem, in that I could see the
complete admin page that was previously truncated but now nothing works - I
get internal server error for all pages on the server.
Regards, ADN
--
You received this message because you are subscribed to the Google Groups
"Django users" group.
To view this discussion on the web visit
https://groups.google.com/d/msg/django-users/-/TfKyWuSl_TkJ.
To post to this group, send email to django-users@googlegroups.com.
To unsubscribe from this group, send email to
django-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/django-users?hl=en.
