It's interesting you mention this. I noticed something with OrderPayment
recently too. I thought it was just the way I was generating the documents
but maybe there is something going on here.

I'm wondering if there was a regression introduced recently. One issue might
be trying to roll back a ways and see if the error still occurs.

-Chris

On Thu, Jan 6, 2011 at 10:32 AM, John-Scott
<[email protected]>wrote:

> I recently migrated 6 sites to a new host (running
> nginx/apache-mpm-worker/modwsgi 3.3 daemon-mode on Ubuntu 10.4 LTS). They
> all share the same project directory and virtualenv but each have their own
> settings file and wsgi config. Only 3 are running satc...@fed2d6d6d239,
> but aside from Satchmo and its dependencies, all other apps are identical
> across the sites. The old vm was essentially the same, just Ubuntu 9.10 with
> a Satchmo checkout from October.
>
> The issue is that I'm getting very random errors, "ImportError: cannot
> import name OrderPayment" (see sanitized stack trace [1]).
>
> This error is extremely random (if I had to guess, 1 in 50-100 requests per
> site) and I cannot consistently reproduce it (only once have I even
> triggered one of these errors in all my prodding).
>
> The summary of what's going on in the traceback:
>
>    - admin.autodiscover() is in the middle of importing the models from
>    satchmo_store.shops.models.
>    - While parsing OrderBasePayment, it then needs to lookup a custom
>    field in payment.fields.
>    - PaymentChoiceField's choices set off a chain of method calls that
>    results in models.get_apps() being called.
>    - models.get_apps() of course starts the model import process over.
>    - While models.get_apps() imports payment.models, it attempts to import
>    OrderPayment from satchmo_store.shops, which fails for reasons not clear to
>    me.
>
>
> Any ideas of what could be causing this random error? Or tips on how to
> troubleshoot this more effectively?
>
> It seems if this was purely an import error, it would always be an error
> and not just *sometimes*. At this point it does seem to be some odd
> combination of Satchmo (or one of its dependencies) and the particulars of
> my environment since the 3 non-Satchmo sites are working flawlessly. Any
> keyedcache or threadlocal wonkiness that might be wreaking havoc?
>
> Thanks,
> John-Scott
>
> [1] http://dpaste.com/hold/293523/
>
> --
> 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]<satchmo-users%[email protected]>
> .
> For more options, visit this group at
> http://groups.google.com/group/satchmo-users?hl=en.
>

-- 
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.

Reply via email to