|
From: Dirk-Willem v. G. <di...@we...> - 2015-04-07 16:19:38
|
> On 07 Apr 2015, at 18:05, Ludovic Rousseau <lud...@gm...> wrote: > > 2015-04-03 15:50 GMT+02:00 Dirk-Willem van Gulik <di...@we...>: >> >> On 03 Apr 2015, at 15:06, Ludovic Rousseau <lud...@gm...> >> wrote: >> >> As it is not uncommon to do roll-over smartcards; which already have the >> ‘new’ cert or still have the expired cert. >> >> >> I changed the "break;" in "continue;" in >> https://github.com/OpenSC/pam_pkcs11/commit/4ef003ac43405f6391bf965a043f9fe4c4704f1d >> >> >> I think that is not quite enough - as those breaks where just breaking out >> of the switch; not out of the encapsulating loop. > > Exact. I reverted the patch. > >> Secondly - when NSS is in use - there is the issue that one cannot >> distinguish between normal invalid cert errors and nefarious/buggy errors >> within NSS/OpenSSL. >> >> So I guess at first oder we need something like: >> >> https://github.com/dirkx/pam_pkcs11/commit/f6a0b89d691b560ea546cab1bd74bc68c2dc1df5 >> >> and at second order - we could consider pulling line 670 onwards into the >> loop as well. To really try them one by one. > > Please create a Pull Request if you consider your code finished. Will do. For now though - not ready/finished. However, 建明, it may be of use to you — as it solves the dominant case; of having multilple keys on one card; only *one* (or the last one) of which should be tried (count toward any PIN entry, etc). > I can also create a team of maintainers for pam_pkcs11 and add > volunteers to the team. Do you volunteer? Let me check some i’s and cross some legal t’s first. Will let you know, Dw. |