I had this exact problem with `OrderPayment`. It only occurred when importing the `Order` model class in my own module (which was last in the `INSTALLED_APPS` settings, and only when running apache+wsgi in my deployment environment.
Your modification of payment/models.py to load `OrderPayment` lazily fixed the problem. On Jan 16, 4:24 pm, Jakob H <[email protected]> wrote: > Hi, > > Thanks for the feedback. I tried Hynek's method, but it didn't help. > The only module that wasn't loaded was 'tax.modules.no' but the > provided code snipped didn't help from what I could see. > > What seems to have fixed it for now at least was the following > modification to satchmo/apps/payment/models.py: > > 1. Comment out "from satchmo_store.shop.modelsimportOrderPayment" > from imports at top of the file (line 15). > 2. Do lazy loading of OrderPayment on line 46: > > orderpayment = models.ForeignKey('OrderPayment', unique=True, > related_name="creditcards") > > Other comments: > > 1) In my satchmo.log file I see the following: > > SettingNotSet: SHIPPING.MODULES > SettingNotSet: PAYMENT_PAYPAL.CREDITCHOICES > > Not entirely sure what they refer to or if they are related in any > way. > > 2) I never see the error when running runserver on my local machine. > > 3) On my local machine running Apache through WSGI, when I run the > WSGI script (./django.wsgi) I don't get theimporterror > > 4) However, on my deployment server (shared hosting) when I run my > WSGI script (./django.wsgi) I get theimporterror every time (without > my above-mentioned 'fix') > > I do not know what the difference is between my local machine (running > Apache) and my shared hosting service that causes the difference in > results. It's the same code, and I've tried to replicate the version > of every library that I know I use and use the same version of > Python. > > I'm not really an experienced Python developer so this is all I know > for now. Anyway, it seems to be the cause of a very specific issue of > my deployment configuration, but I don't know what. And, at least the > above-mentioned fix has solved it for me for now. > > Again, thanks for your help, > Jakob > > On Jan 13, 12:43 pm, John-Scott Atlakson > > > > > > > > <[email protected]> wrote: > > Hi Jakob, > > > This rearranging of imports solved a (separate?) issue I was seeing where > > OrderTaxDetail could not be imported. This had the mysterious effect of > > eliminating the large number of OrderPaymentimporterrors, evenerrorsI > > was seeing in sites using an unmodified Satchmo in a separate virtualenv. > > > I think Chris M was suggesting that you try Hynek's strategy mentioned > > above. > > > One simple thing would be to comment out admin.autodiscover() and see if the > > problem goes away. That doesn't tell us much, but if this is a live site > > that you need to get back up in a rush, this may be a stop gap. > > > I may have to take part of this conversation over to the modwsgi group, but > > there is definitely something specific about Satchmo and/or some of its > > depencies (keyed-cache, threaded-multihost, etc) since I do not have this > > problem with any other non-Satchmo projects (even on the same server using > > the same virtualenv, just no satchmo in INSTALLED_APPS). > > > Cheers, > > John-Scott > > > On Thu, Jan 13, 2011 at 11:03 AM, Jakob H <[email protected]>wrote: > > > > Hi, > > > > Not yet, but I will try this tonight. So, the change I will try is: > > > > 1. Move class definition of OrderTaxDetail above class definition of > > > Order > > > 2. Put the FK model reference (Order) within single quotation marks > > > ('Order') to make it load lazily (did not know you could do that) > > > > Does this seem correctly understood? > > > > I will try it and let you know. > > > Cheers, > > > Jakob > > > > On Jan 13, 10:34 am, Chris Moffitt <[email protected]> wrote: > > > > Did you try the solution that was posted earlier in this thread? I > > > haven't > > > > tried it yet to see if it fixes it for me. > > > > > -Chris > > > > > On Thu, Jan 13, 2011 at 9:15 AM, Jakob H <[email protected]> > > > wrote: > > > > > Hi, > > > > > > I now get this error on basically every request to the website. > > > > > However, I'm not sure it is specifically Satchmo related, but I really > > > > > need to get this fixed. > > > > > > The problem does not seem to occur on the runserver on my local > > > > > machine, but does happen every time on my shared host deployment > > > > > server using WSGI. > > > > > > Here is my traceback:http://dpaste.com/313255/. > > > > > > I have Python 2.6.1 on my local machine and Python 2.6.2 on my > > > > > deployment server if that matters. > > > > > > Cheers, > > > > > Jakob > > > > > > On Jan 12, 10:20 pm, John-Scott <[email protected]> wrote: > > > > > > Thanks for your thoughtful and thorough reply, Hynek. > > > > > > > Unfortunately I have been swamped by other projects for the moment > > > and > > > > > > haven't had a spare moment to rigorously troubleshoot (fortunately > > > the > > > > > >errorsare rare and as mentioned below do not seem to be triggered by > > > > > human > > > > > > visitors). > > > > > > > As mentioned in my other thread [1], my issues have* mostly > > > *vanished. I > > > > > > still occasionally see animporterror, although much less > > > frequently. > > > > > For > > > > > > instance, I had to restart Apache and memcached recently and the > > > first > > > > > > requests to my 3 Satchmo sites each generated a single error email > > > with > > > > > the > > > > > >importerror as the cause. After that, everything was fine. I've also > > > > > seen > > > > > > morerandomerrorswhere searchbots hit a completely invalid url on > > > my > > > > > sites > > > > > > (scratching my head where they've come up with .aspx and .htm urls > > > > > > on > > > a > > > > > site > > > > > > that has only been Django based and has only had Django-style > > > 'pretty' > > > > > > urls). Don't know that it's cause/effect...sadly the majority of > > > traffic > > > > > is > > > > > > probably from searchbots to begin with, but the error messages > > > > > > always > > > > > have > > > > > > googlebot, bingbot or Yahoo! Slurp (gross) as the user agent and the > > > > > > majority of the time they are hitting urls that have never been > > > valid. > > > > > > > Again, haven't hide the time to do another deep dive, but thought > > > > > > I'd > > > > > toss > > > > > > these anecdotes onto the pile. > > > > > > > Cheers, > > > > > > John-Scott > > > > > > > [1] > > > > >https://groups.google.com/d/topic/satchmo-users/4nMkBFw7oGQ/discussion > > > > > > -- > > > > > 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%2Bunsubscribe@goog > > > > > legroups.com> > > > <satchmo-users%2Bunsubscribe@googlegroups.com> > > > > > . > > > > > 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]<satchmo-users%2Bunsubscribe@goog > > > legroups.com> > > > . > > > 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.
