On Tue, 2016-12-13 at 14:42 -0800, Mike Miller wrote: > Wow, thanks, after simply copying tests/softhsm2.conf to > tests/.config/softhsm2/softhsm2.conf the test passes. Good find, I was > not seeing this in my strace for some reason. > > Would you be open to using .config/softhsm2 upstream or do you want to > keep using the environment variable? Hm, I'm not convinced that $HOME/.config/softhsm2/ thing is upstream; is it a Debian patch? It makes some sense; get it upstream and I'll chase it into Fedora too. I suppose putting it in $HOME/.config/softhsm2/ and then using the (upstream) SOFTHSM2_CONF environment variable to point to the same place should work for everyone, and I can't see a problem with that in the short term. Why *doesn't* the Debian package honour $SOFTHSM2_CONF though? > Any thoughts about how to make the contents of softhsm2.module more > distro-agnostic? Build it (from the distro's own one) at configure time? We already build softhsm2.conf from softhsm2.conf.in anyway. > Btw, the setrlimit error was on the ocserv side, and can be safely > ignored, at least for the purposes of this test suite. OK. > > Can I leave you to work out what needs fixing in the packaging of > > SoftHSM and what needs fixing in OpenConnect? > > Sure, I'm not directly involved, but I'll try to push for some fixes for > these configuration issues. Thanks. Might be worth keeping an eye on this one which bites OpenConnect too: https://bugs.launchpad.net/ubuntu/+source/gnutls28/+bug/1649270 -- dwmw2 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5760 bytes Desc: not available URL: <http://lists.infradead.org/pipermail/openconnect-devel/attachments/20161213/0664c0d3/attachment.bin>
