[ic] tag_data() triggers Sig11 on Cygwin w/ DBI

[interchange-users@interchange.redhat.com (Dan Browning)]

I just posted bug 409 
(http://interchange.redhat.com/bugs/show_bug.cgi?id=409), but I thought I 
would attach it here as well, in the hopes that someone might pipe up with 
some help.

Thanks,

+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Dan Browning, Kavod Technologies <db@kavod.com>
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Real programmers don't bring brown-bag lunches.  If the vending machine
doesn't sell it, they don't eat it.  Vending machines don't sell
quiche.

#####################################
##   Environment
#####################################

* Cygwin latest (3-20-02)
* IC 4.8.4 cvs 3-21-02 & Bundle::Interchange "up to date"
* Foundation catalog using DBI
  - Tried mysql *and* pgsql to no avail.
  - (GDBM works perfectly)

#####################################
##   Problem
#####################################

The checkout and Admin->Layout pages cause this error:

Had to create DBI::_dbistate unexpectedly at (eval 271) line 3.
Signal 11

#####################################
##   Steps to reproduce
#####################################

This snippet of code from checkout.html...

[perl tables="country"]
         return tag_data('country', 'shipmodes', 'US');
[/perl]

Produces this error message...

Had to create DBI::_dbistate unexpectedly at (eval 271) line 3.
Signal 11

(It correctly returns "1DA 2DA GNDCOM" when using GDBM instead of DBI).

#####################################
##   Conclusion
#####################################
It's probably a bug in the DBI code, but it would be great if there was a
workaround so that tag_data() wouldn't trigger it.  Thanks!