Paride Legovini
2010-Jan-02 18:17 UTC
pkg_add thinks that -STABLE is -RELEASE (wrong __FreeBSD_version?)
Hi, I'm running 8.0-STABLE and I noticed that by default pkg_add -r uses packages-8.0-release/Latest/ as PACKAGESITE. I read in the handbook[1] that pkg_add should use packages-5-stable/Latest/ as PACKAGESITE when one is running -STABLE. I took a look at the source code, and noticed that pkg_add considers the system -STABLE if getosreldate() (i.e. __FreeBSD_version) is between 800500 and 899000 (see src/usr.sbin/pkg_install/add/main.c:92). However, in RELENG_8, __FreeBSD_version is set to 800108. You can check it via cvsweb[2]. Sounds like something is wrong. Am I missing something? Do I just have to wait for the __FreeBSD_version to be bumped? Regards, Paride [1] http://www.freebsd.org/doc/en/books/handbook/packages-using.html [2] http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/sys/param.h?only_with_tag=RELENG_8
pluknet
2010-Jan-02 21:20 UTC
pkg_add thinks that -STABLE is -RELEASE (wrong __FreeBSD_version?)
2010/1/2 Paride Legovini <pl@ninthfloor.org>:> Hi, > > I'm running 8.0-STABLE and I noticed that by default pkg_add -r uses > packages-8.0-release/Latest/ as PACKAGESITE. I read in the handbook[1] > that pkg_add should use packages-5-stable/Latest/ as PACKAGESITE when > one is running -STABLE. > > I took a look at the source code, and noticed that pkg_add considers > the system -STABLE if getosreldate() (i.e. __FreeBSD_version) is between > 800500 and 899000 (see src/usr.sbin/pkg_install/add/main.c:92). > However, in RELENG_8, __FreeBSD_version is set to 800108. You can check > it via cvsweb[2]. > > Sounds like something is wrong. Am I missing something? > Do I just have to wait for the __FreeBSD_version to be bumped?Hi. I'm afraid that's because __FreeBSD_version wasn't bumped 800107->800500 in RELENG_8 just after RELENG_8_0 created (wrt changes in scheme for RELENG_8 timeframe where current/stable border moved to 800500: http://lists.freebsd.org/pipermail/svn-src-head/2009-June/007830.html It continued then as is (still ok), and eventually was incremented to 800108 (wrong here, though I hope it still can be safely corrected). -- wbr, pluknet