Interesting. I am currently taking a look at this. As far as I understood then, the current packages for RHEL/CentOS should be ok, but those for Debian and Ubuntu are not. Is this correct so far?
2015-03-12 15:13 GMT+01:00 Michael Biebl <mbi...@gmail.com>: > An more advice: > Please always consider running lintian on the final package. > > This would have caught simple errors like this one: > > $ lintian liblognorm1_1.1.1-0adiscon1trusty1_amd64.deb > W: liblognorm1: package-name-doesnt-match-sonames liblognorm2 > E: liblognorm1: > symbols-file-contains-current-version-with-debian-revision on symbol > descent_parser_data_constructor@Base and 73 others > > It looks, like the symbols file isn't properly maintained either. > > 2015-03-12 15:10 GMT+01:00 Michael Biebl <mbi...@gmail.com>: > > Apparently, the liblognorm soname bump wasn't properly handled > > packaging wise looking at > > > > https://launchpad.net/~adiscon/+archive/ubuntu/v8-stable/+packages > > > > Packaging (and maintaining) libraries takes special effort to be done > properly. > > > > Please have a look at > > https://www.debian.org/doc/debian-policy/ch-sharedlibs.html > > https://www.debian.org/doc/manuals/maint-guide/advanced.en.html > > https://www.netfort.gr.jp/~dancer/column/libpkg-guide/libpkg-guide.html > > > > 2015-03-12 15:05 GMT+01:00 Michael Biebl <mbi...@gmail.com>: > >> I do hope, the Debian packages were updated accordingly, i.e. if there > >> is a soname bump, you name it liblognorm2. > >> So if you rebuild rsyslog against liblognorm-dev 1.1.1, it would get a > >> dependency on liblognorm2. This way, it's ensured that you can't end > >> up with broken packages. > >> > >> 2015-03-12 14:55 GMT+01:00 Tomas Heinrich <thein...@redhat.com>: > >>> On 03/12/15 13:07, Thomas D. wrote: > >>>> > >>>> Rainer Gerhards wrote: > >>>>> > >>>>> that probably has to do with the fact that we fixed the library > >>>>> versioning. > >>>>> I admit we did it wrongly so far, so I have little experience with > the > >>>>> proper way and its implications. I guess, we need to rebuild the > rsyslog > >>>>> packages. Will initiate that. > >>>> > >>>> > >>>> Right, you must rebuild against the the SO name version you are using. > >>> > >>> > >>> Yes, that needs to be done every time the value of > >>> first-digit-minus-third-digit of the soname changes. > >>> > >>>> Don't forget to require liblognorm-1.1.1 in your packages if you build > >>>> them against liblognorm-1.1.1 because if you link against > >>>> liblognorm-1.1.1 you can't use previous versions anymore (or you have > to > >>>> rebuild...). > >>> > >>> > >>> Actually, at least for RPM, that is not necessary, generally speaking. > >>> During the build, dependency on a specific soname is automatically > recorded > >>> in the depending package: > >>> > >>> $ rpm -qR rsyslog-mmnormalize | grep liblognorm > >>> liblognorm.so.2()(64bit) > >>> > >>> Upstream tracker shows[0] that 1.1.1 is backwards compatible to 1.1.0, > so > >>> unless you need the symbols that were added in 1.1.1, versioning the > >>> dependency in the package shouldn't be required. > >>> > >>> (And looking at the actual soname - 2:0:0 - since the "age" part wasn't > >>> incremented, the library will be considered as not backwards > compatible and > >>> the requirement has to be placed on the newest version.) > >>> > >>> Tomas > >>> > >>> http://upstream.rosalinux.ru/versions/liblognorm.html > >>> > >>> > >>> _______________________________________________ > >>> rsyslog mailing list > >>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>> http://www.rsyslog.com/professional-services/ > >>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > myriad of > >>> sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > DON'T > >>> LIKE THAT. > >> > >> > >> > >> -- > >> Why is it that all of the instruments seeking intelligent life in the > >> universe are pointed away from Earth? > > > > > > > > -- > > Why is it that all of the instruments seeking intelligent life in the > > universe are pointed away from Earth? > > > > -- > Why is it that all of the instruments seeking intelligent life in the > universe are pointed away from Earth? > _______________________________________________ > rsyslog mailing list > http://lists.adiscon.net/mailman/listinfo/rsyslog > http://www.rsyslog.com/professional-services/ > What's up with rsyslog? Follow https://twitter.com/rgerhards > NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad > of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > DON'T LIKE THAT. > _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
