Control: severity -1 important
I've been trying some regression tests [removing the new CONFARGS in
debian/rules, removing disable_sslv3_test.patch] -- and eventually tried
rebuilding openssl_1.0.2d-1 as a check and I see the same issue after
building Mumble with it. In other words: re-compiling the version of
openssl + mumble prior to the bug now exhibits the bug too.
It looks now like the changes to the openssl_1.0.2d-3 package aren't the
cause of the behavior change.
The logs for openssl builds show some differences since openssl_1.0.2d-1 was
last built in July -- new packages pulled in:
dh-strip-nonteterminism
libfile-stripnondeterminism-perl
libtimedate-perl
libarchive-zip-perl
[These get pulled in when building openssl_1.0.2d-1 now too,
where they didn't at the time it was built on the buildds.]
And of course gcc/g++ 5.2.1 is used vs gcc 4.9.3 previously.
I'd like to try building openssl_1.0.2d-1 with gcc-4.9 to see if the
behavior change was due to gcc-5 -- if there's an easy way to specify that
please let me know.
Thanks.
-- Chris
--
Chris Knadle
chris.kna...@coredump.us
