>>>>> "Rich" == Richard Lowe <richl...@richlowe.net> writes:
Rich> it smells of mismatched closed-bins to me.
Give the man a prize. Yes, the b111 closed bins tarball appears to have
a post-111 pkcs11_softtoken.so.1. It doesn't have other closed binaries
from snv_112 that I could see (I checked the glm driver and libike).
pkcs11_softtoken.so.1 is one of those magic signed crypto files, isn't
it?
Here's a relevant section from bindrop.sh (which generates the closed
binaries tarball):
#
# Crypto related binaries need to be signed in order to be loaded.
# We pull the ongk signed binaries from the gate machine's build
# at the path below so that the closed-bins tarballs are kept in sync
# with what we're actually delivering. We default to pulling out of
# nightly, but if CRYPTO_BINS_PATH is set, then we pull from that path
# instead. This allows us to override with something like
# /ws/onnv-gate/packages/$isa/snv_XX instead.
#
gatepkgs=${CRYPTO_BINS_PATH:-"/ws/onnv-gate/packages/$isa/nightly"}
So if, for example, CRYPTO_BINS_PATH was not set when the closed bins
were generated, bindrop would pick up the latest nightly packages.
Dina's putback was on 20 March, and the closed bins were generated on 25
March.
BTW, for those following at home, the CR is
6824969 snv_111 nightly fails to build from source archive
mike
_______________________________________________
opensolaris-code mailing list
opensolaris-code@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
