Created attachment 574616 [details] dev-ruby:facter-3.13.2:20190429-153919.log.bz2 This is subslot rebuild post boost upgrade, so it seems that subslot machinery somehow doesn't get it right. /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: warning: libboost_regex.so.1.65.0, needed by /usr/lib64/leatherman_curl.so.1.6.0, may conflict with libboost_regex.so.1.70.0 /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: warning: libboost_filesystem.so.1.65.0, needed by /usr/lib64/leatherman_curl.so.1.6.0, may conflict with libboost_filesystem.so.1.70.0 /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: warning: libboost_locale.so.1.65.0, needed by /usr/lib64/leatherman_locale.so.1.6.0, may conflict with libboost_locale.so.1.70.0 /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: warning: libboost_chrono.so.1.65.0, needed by /usr/lib64/libboost_log.so.1.65.0, may conflict with libboost_chrono.so.1.70.0 /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: warning: libboost_thread.so.1.65.0, needed by /usr/lib64/libboost_log.so.1.65.0, may conflict with libboost_thread.so.1.70.0 /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../lib64/libcpp-hocon.a(parseable.cc.o): undefined reference to symbol '_ZN5boost6detail12set_tss_dataEPKvNS_10shared_ptrINS0_20tss_cleanup_functionEEEPvb' /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: /usr/lib64/libboost_thread.so.1.65.0: error adding symbols: DSO missing from command line collect2: error: ld returned 1 exit status
do you get this with boost 1.65? it could be that facter just doesn't support boost 1.70.
How could I get it with 1.65 if I'm clearly saying it happens due to the upgrade? Apparently its deps were not rebuilt with boost upgrade. I'm going to go through them and add := whenever appropriate.
What I'm really wondering about is why Portage didn't rebuild leatherman before facter but I don't really have the patience to try to figure it out.
I'm adding := to cpp-hocon, and after rebuilding cpp-hocon and leatherman, facter finally builds.