Missed shipping charge

[ids]

The store is a customized (not very much) Colossus RT in the store and no customization to shipping modules. And, I have not been able to reproduce the scenario.

What Happened?
A customer put items in the basket and didn't qualify for free shipping. The customer chose the Priority Next Day likely because the price showed up incorrectly at $0.00 and remained that way through checkout.

As mentioned, we could not reproduce. The basket always refreshed properly for our testing. (some older 3rd modes won't refresh the basket in case that mattered).

We won't get to know what rendered for the customer when they thought they were getting a great deal on shipping. And, I can't find a trigger for the wrong price so I can't really get creative to try and prevent this from happening again. Any ideas on how shipping integrity could be maintained through checkout? It's an anomaly of sorts, but, it's pretty hard to do that in this software.

Scott

[Bruce - PhosphorMedia]

The only thing i can think of is somehow the shipping methods array had a member that was missing the price data. Long shot, but see if you can find a scenario where every available shipping method is queried and then examine the array?

[ids]

I'd actually thought about adding some template code test price value of the shipping options presented. It's in a foreach loop already. But, not sure what to do if the wrong values show up. Presumably, the customer has nothing to do with it, so, would be confusing for a customer to see an error of some sort.

Scott

[Bruce - PhosphorMedia]

If anything, write the stuff to a log and then set a reminder to look at it in week :)

BTW, what shipping methods are being used and are there any special rules? I've gotten into situations where shipping rules unexpectedly collide, though that always resulted in the "methods not found" error...still there are so many rules, that this might be a possibility.

[ids]

Along with no real customizations, there are no shipping rules that would potentially be in a conflict of some sort. This was a first and haven't been able to reproduce.

The testing and logging seem to be likely the next step unless someone else can provide their experiences.

Scott