Hello Alex, in my attempt to get some details out of it I needed these steps: - connection to a MacOSX (Apple remote desktop) - enable fips mode - reset initialized in basic_initialization - reset thread_model in _gcry_ath_mutex_init
Points 3 and 4 I had to change in debugger. Therefore I would still classify it as unreproducible. Right now I could only imagine that fips mode is really activated at the sumitters system and for some reason either random number generator gets not initalized properly or some memory gets overwritten. Both cases I cannot confirm or deny as on my system no crash happens (without me forcing it to). Some answers from the submitter would be highly helpful: - is it still happening with current versions? - is fips mode enabled? - are some unusually long computer name, user name or password involved? If no other people can reproduce it and there is no response from the submitter received, then probably the justification "renders package unusable" is a little too much? Kind regards, Bernhard PS.: What happens with release critical bugs which are not reproducible and where the submitter does not respond? (for my part, not being maintainer or otherwise involved into vinagre or gtk-vnc, when we not get response from submitter then decreasing severity to normal would be adequate?) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
