Make errors

[patrickminas]

Im on CentOS 6.5 x86_64. Fresh installation and fully updated. Im getting this at the end of the make.

zone_data.cc: In member function ‘void bundy::datasrc::memory::ZoneData::removeNode(bundy::util::MemorySegment&, bundy::datasrc::memory::ZoneNode_)’: zone_data.cc:265: error: no match for ‘operator==’ in ‘node == ((bundy::datasrc::memory::ZoneData_)this)->bundy::datasrc::memory::ZoneData::origin_node_’ make[8]: *** [zone_data.lo] Error 1 make[7]: *** [all-recursive] Error 1 make[6]: *** [all] Error 2 make[5]: *** [all-recursive] Error 1 make[4]: *** [all] Error 2 make[3]: *** [all-recursive] Error 1 make[2]: *** [all-recursive] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2

Configure command: [root@dns1 bundy]# ./configure --prefix=/opt/bundy10 --with-pythonpath=/opt/python3/bin/python3 --with-botan-config=/opt/botan/bin/botan-config-1.10 --with-log4cplus=/opt/log4cplus/ --without-shared-memory

[jinmei]

thanks for the report. I believe commit f840e89 should fix that. could you pull it and try again?

[patrickminas]

getting those errors when running make

cc1plus: warnings being treated as errors ../../../../src/lib/dns/serial.h: In member function ‘bundy::datasrc::memory::ZoneDataLoader::LoadResult bundy::datasrc::memory::ZoneDataLoader::ZoneDataLoaderImpl::doLoadCommon(bundy::datasrc::memory::detail::SegmentObjectHolder<bundy::datasrc::memory::ZoneData, bundy::dns::RRClass>, bundy::datasrc::memory::ZoneDataLoader::ZoneDataLoaderImpl::RRsetInstaller)’: ../../../../src/lib/dns/serial.h:75: error: dereferencing pointer ‘’ does break strict-aliasing rules zone_data_loader.cc:516: note: initialized from here make[8]: ** [zone_data_loader.lo] Error 1 make[8]: Leaving directory /root/bundy/src/lib/datasrc/memory' make[7]: *** [all-recursive] Error 1 make[7]: Leaving directory/root/bundy/src/lib/datasrc/memory' make[6]: *** [all] Error 2 make[6]: Leaving directory /root/bundy/src/lib/datasrc/memory' make[5]: *** [all-recursive] Error 1 make[5]: Leaving directory/root/bundy/src/lib/datasrc' make[4]: *** [all] Error 2 make[4]: Leaving directory /root/bundy/src/lib/datasrc' make[3]: *** [all-recursive] Error 1 make[3]: Leaving directory/root/bundy/src/lib' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory /root/bundy/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory/root/bundy' make: *** [all] Error 2

[jinmei]

hmm, I remember we previously encountered odd g++ errors like this with boost optional. Doesn't this something like this? https://bugzilla.redhat.com/show_bug.cgi?id=531483. If so, and if you can, upgrading g++ or using a different compiler might help. Alternatively, if your goal is to just make the build pass, you could use the --without-werror ./configure option.

That said, even if this is a compiler bug, the sad reality is that application developers often need to work around it themselves, especially if it's a widely used compiler. If this is one of such cases, the change on branch 'issue39' (or applying commit 06c1aad by hand) I just pushed to github could be used. If you''re willing to, please try these and let me know which/whether they work.

Thanks.

[patrickminas]

I am not a developer. I just want to install Bundy DNS on a fresh CentOS 6.5 Server.

Will it be possible to remotely connect and check the issue urself? if yes, we can chat on skype and coordinate (skype: patrick_m_68)

On Jun 8, 2014, at 3:17 AM, JINMEI Tatuya [email protected] wrote:

hmm, I remember we previously encountered odd g++ errors like this with boost optional. Doesn't this something like this? https://bugzilla.redhat.com/show_bug.cgi?id=531483. If so, and if you can, upgrading g++ or using a different compiler might help. Alternatively, if your goal is to just make the build pass, you could use the --without-werror ./configure option.

That said, even if this is a compiler bug, the sad reality is that application developers often need to work around it themselves, especially if it's a widely used compiler. If this is one of such cases, the change on branch 'issue39' (or applying commit 06c1aad by hand) I just pushed to github could be used. If you''re willing to, please try these and let me know which/whether they work.

Thanks.

— Reply to this email directly or view it on GitHub.

[jinmei]

okay, in that case I recommend you to just suppress the warning. Try % ./configure --without-werror < other options, if any> and let me know if the build still fails.

[jinmei]

hmm, somehow this one was also closed, but that wasn't my intent...anyway, I believe the build issue was solved on the latest master. Could you pull it and try again?