I would start digging at the source of the error chain:
> > File "/var/www/vhosts/server_name/python/lib64/python2.5/site-
> > packages/django/db/models/sql/query.py", line 1734, in
execute_sql
> > cursor.execute(sql, params)
> > File "/var/www/vhosts/server_name/python/lib64/python2.5/site-
> > packages/django/db/backends/util.py", line 19, in execute
> > return self.cursor.execute(sql, params)
Edit the .../util.py file and insert a print sql; print params statement
line. The you take a look how the combined sql would look like and you will
find maybe some strange artifacts.
If you find, then the reason why it was no problem with sqllite was maybe,
that is was just more forgiving in this case. And, you might try to
investigate what is the source for the bad statement - bad data in params
might be a good guess. ...
Hth G
-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]]
Im Auftrag von rajeev j sebastian
Gesendet: Donnerstag, 08. Oktober 2009 19:55
An: Satchmo users
Betreff: Re: satchmo +nginx+fcgi
On Oct 8, 10:23 pm, Chris Moffitt <[email protected]> wrote:
> Do you see anything in your postgres logs that might point to the issue?
>
> Do other Postgres databases work fine? I might start a very simple django
> project and try to connect. I'm inclined to believe it's not a
> Satchmo-specific problem.
Maybe it is, maybe it isnt. Can't be sure whether its Satchmo or not.
These are the things I can see:
1) works on sqlite+flup+nginx (for many weeks)
2) does not work on PG+flup+nginx
3) *does* work on PG+devserver
It seems that for some reason, some db access is done after the db
connection is closed. This is why the exception happens. As for why it
happens, I'm not quite sure. I use runfcgi with prefork,
maxrequests=1, debug=true and maxchildren=1 for testing. Flup
traceback is either not very interesting or very hard to parse :)
Others have posted the same issue over the months. See another example
here:
http://osdir.com/ml/DjangoUsers/2009-04/msg01647.html
See also Malcolm Tredinnick's comment.
He was suggesting to go through the entire reqeust-response cycle to
figure out where this happens. So far, I'm not sure how I could do
that.
I found some places in satchmo code where livesettings were being
queried for building url patterns (e.g., the product and category base
slugs). I removed them for testing.
Same thing for a couple of the Middlewares provided by satchmo.
Still no luck.
Any pointers ?
Regards
Rajeev J Sebastian
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Satchmo users" 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/satchmo-users?hl=en
-~----------~----~----~----~------~----~------~--~---
