Dear Maintainer,
ruby-odbc does this, at least for me, on Buster:
martind@neutrino:~$ ruby -we 'require "odbc"; ODBC.drivers()'
Traceback (most recent call last):
1: from -e:1:in `<main>'
-e:1:in `drivers': INTERN (0) [RubyODBC]Cannot allocate SQLHENV (ODBC::Error)
martind@neutrino:~$
The same test on the same configuration on Stretch is silent:
martind@scoot:~$ ruby -we 'require "odbc"; ODBC.drivers()'
martind@scoot:~$
I do have non-default /etc/odbc{,inst}.ini files but removing them doesn't affect this test.
Buoyed by a mention of Stretch here:
https://bugs.launchpad.net/raspbian/+bug/1832778
... I thought that maybe this blast from the past could have come back:
http://opensysblog.directorioc.net/2013/07/rubyodbccannot-allocate-sqlhenv.html
The change log in:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=889025
... included:
* Refresh 001extconf_dlopen.patch
Reverting that fixed the problem.
These are the steps I took:
aptitude source ruby-odbc
sudo aptitude install gem2deb
tar --xz -xf ruby-odbc_0.99998-1.debian.tar.xz
debian/patches/ext_enable_dlopen.patch
cd ruby-odbc-0.99998/
patch -p1 -R < ../debian/patches/ext_enable_dlopen.patch
dpkg-buildpackage -b -nc -uc
sudo dpkg -i ../ruby-odbc_0.99998-1_amd64.deb
I can't say I understand why Debian would have to disable dlopen here.
But the package appears unusuable without that, for me.