Todd Denniston wrote:
I have not tried that yet (locked my card with some of the tests yesterday), but looking at the changes in the patch still applied (item 7 from Andrew Pimlott[1]) I don't see any thing that should be messing with the data that is causing the problems.
For the record, with the subset of the patches described by Andrew that were applied in SmartCardServices-26726 (and 26777), libmusclepkcs11.so behaves correctly in Firefox. Prior to this version (and without any of the patches Andrew describes), Firefox could open the card and view certificates but could not complete client SSL authentication; this appears to be fixed by the first chunk.
The issue I'm seeing (segfault during session close) is killing me with pam_pkcs11, but I also saw it with Geoff's test code--although today that code is behaving differently; now pcscd is reporting that the token isn't present when session-test opens the session. Odd that it changed.
Everything on my end seems to be pointing to a bug with token status checking when commonAccessCard.so is loaded. FWIW I've tried multiple readers.
Interestingly, when pcscd starts I'm seeing occasional errors: there's a broken pipe error during get_data_rates() followed by a usb_bulk_read error, IFDStatusICC() Card not transacted error and EHSpawnEventHandler() failure. Not sure what's happening there but it may be related.
-- Tim
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ Muscle mailing list [email protected] http://lists.drizzle.com/mailman/listinfo/muscle
