Date: Wed, 23 Sep 2020 15:53:01 +0200 From: Baptiste Daroussin <bapt@FreeBSD.org> To: Matthias Andree <mandree@FreeBSD.org> Cc: ports-committers@freebsd.org, svn-ports-all@freebsd.org, svn-ports-head@freebsd.org Subject: Re: svn commit: r549723 - head/sysutils/e2fsprogs Message-ID: <20200923135301.b4y7bcgyuawhvvy2@ivaldir.net> In-Reply-To: <202009231244.08NCiHUQ090692@repo.freebsd.org> References: <202009231244.08NCiHUQ090692@repo.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--l4aj7xlworhkp6wn Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Sep 23, 2020 at 12:44:17PM +0000, Matthias Andree wrote: > Author: mandree > Date: Wed Sep 23 12:44:16 2020 > New Revision: 549723 > URL: https://svnweb.freebsd.org/changeset/ports/549723 >=20 > Log: > - make /sbin/e2fsck and /sbin/fsck_ext2fs visible in pkg catalog/"pkg w= hich", > so that "pkg which /sbin/e2fsck" yields the proper result. > =20 > * this entails symlinking from $PREFIX/sbin to /sbin, and the pkg-ins= tall > script will attempt to replace the symlinks by hard links if possib= le. > pkg 1.15.4 cannot deal with packaged hard links and will fail. > =20 > * Note that it is unavoidable that these be in /sbin because /usr/loc= al or > /usr may not be mounted and consequently ext2 file systems could not > be fsck-ed or mounted from /etc/fstab. > =20 > There will be no port option to avoid /sbin installs for now. We have= too > many options already and the testing effort increases exponentially. > =20 > - make sure pkg-message appears on both install and upgrade > =20 > - clean up and document/comment pkg-install so that armchair experts > will not pester me with meaningless change requests > =20 > - bugfix/change: logic of mke2fs.conf upgrade handling to present less = work for > users on port/package upgrades > =20 > - bump PORTREVISION > =20 > PR: 249284 (related) >=20 > Modified: > head/sysutils/e2fsprogs/Makefile > head/sysutils/e2fsprogs/pkg-install > head/sysutils/e2fsprogs/pkg-message > head/sysutils/e2fsprogs/pkg-plist >=20 > Modified: head/sysutils/e2fsprogs/Makefile > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D > --- head/sysutils/e2fsprogs/Makefile Wed Sep 23 12:31:14 2020 (r549722) > +++ head/sysutils/e2fsprogs/Makefile Wed Sep 23 12:44:16 2020 (r549723) > @@ -3,7 +3,7 @@ > =20 > PORTNAME=3D e2fsprogs > PORTVERSION=3D 1.45.6 > -PORTREVISION?=3D 3 > +PORTREVISION?=3D 4 > CATEGORIES?=3D sysutils > MASTER_SITES=3D KERNEL_ORG/linux/kernel/people/tytso/${PORTNAME}/v${PORT= VERSION} > =20 > @@ -252,7 +252,8 @@ post-build: > && ${MAKE_CMD} e2fsck V=3D1 \ > LIBS=3D"-static-libgcc -lc -Bstatic ../lib/libsupport.a ../lib/libext2= fs.a ../lib/libcom_err.a \ > ${_staticlibs} /usr/lib/libexecinfo.a /usr/lib/libelf.a \ > - ${LOCALBASE}/lib/libblkid.a ${LOCALBASE}/lib/libuuid.a ${libintl} ../l= ib/libe2p.a " > + ${LOCALBASE}/lib/libblkid.a ${LOCALBASE}/lib/libuuid.a ${libintl} ../l= ib/libe2p.a " \ > + && ${STRIP_CMD} e2fsck > # Regression check: avoid a port (not upstream!) regression from 1.40.5, > # check that e2fsck isn't dynalinked against anything but libc.so: > @${ECHO_CMD} -n "=3D=3D=3D> checking that e2fsck depends on no shared = objects outside /lib: " > @@ -292,7 +293,24 @@ post-build: > .endif > =20 > post-install: > - ${INSTALL_PROGRAM} ${WRKSRC}/fsck_ext2fs ${STAGEDIR}${PREFIX}/sbin/ > + # pkg workaround time again; This is not a workaround at all, can you please refrain saying those nasty comments and even more commiting them! see below > + # the whole purpose of this dance is to move the necessities for > + # /etc/fstab fsck'ing into /sbin, and symlink to these guys > + # in the regular hierarchy, normally $PREFIX/sbin. Hard links may > + # crash the install if /usr[/local] is a separate file system from /, > + # and pkg 1.15.4 can't automatically unroll or install relative symlink= s. There is no reason why pkg should handle such thing. pkg is supposed to pro= vide always the same setup experience for everyone, so it will not try to create= a link and if it fails copy the file. pkg is supposed to track all files it installs, with special cases for configuration files (@sample) no try to do a weird dance. > + ${MKDIR} ${STAGEDIR}/sbin > + ${MV} -f ${STAGEDIR}${PREFIX}/sbin/e2fsck ${STAGEDIR}/sbin/ # from INST= ALL_TARGET > + ${INSTALL_PROGRAM} ${WRKSRC}/fsck_ext2fs ${STAGEDIR}/sbin/ # from port > + # according to TPH 6.1.1 Handling Symbolic Links, these are to be absol= ute, > + # yet stage-qa warns about these on 2020-09-16: > + ${LN} -fs /sbin/fsck_ext2fs ${STAGEDIR}${PREFIX}/sbin/ > + ${LN} -fs /sbin/e2fsck ${STAGEDIR}${PREFIX}/sbin/ And so instead of helping the framework to improve by providing a patch to stage-qa, or even trying to understand the logic behind those rules you fin= d it better to hard code a comment which complains about it. Bapt --l4aj7xlworhkp6wn Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAABCAAdFiEEgOTj3suS2urGXVU3Y4mL3PG3PloFAl9rUzsACgkQY4mL3PG3 Plpz4g//a8Tedwnn180lnUKub94Ff9c8+vEZuAFcJe27BehHqH/ZL6CgtzwmQvS4 zKYFwUuBK+/CGMuXo7ZJajMpo6SycoMZIVQesDYjJMjNE8LBpVWL8liUVOi0YLB9 Xd570114yNXb3aJ3815r31s9Wjcs8xDszqMv90E46ayYGTJG/wIG0Pdxnf8d/uH4 VzHf+wagmhgO8kRfP/uUDXuhTy+oMM3JDy7pYazIYpK0q/NTTtZDruLUEujcKu2y ewSYjcV4dvkw7iEznS0r6tchSU3cEwJakyamCy/XNgdwzM9Tjf1WFrt7b4+WtOkW AJDexFX8JTgGYYwTELvuiWrHOVVU2eW7PctElXK+BMe9EZXR0+WBE9Y9ecSGAKJA I62h7PptPvXqteSMG2hV6cCSEX87+Z1qYukJGRQUxuRwG4W8j2LUcWMgK5QEkSLF tv3E8osIDLhd9eWEuQf6/hEaObAtZOS9BQ0glKVL04y+1JWFK2rKJQ7Wjw2LB+lS w/p9sLjCFPPa5W8llxVSKxMoXzVfufnUhha+qT0UxQIMTRqEJPAPXqvwl/oZyXwW iQtNEK1MBeVzxNmUfJU7W3FcNLsdQD5b7N91TsCz2ECUBGefadV+w4CO3DYaeZEF M+P2TaEjaaAJ5/fdx+d8Xbp7M8vefXvbEM6PuOfXLgeFWtvfQpA= =m87K -----END PGP SIGNATURE----- --l4aj7xlworhkp6wn--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20200923135301.b4y7bcgyuawhvvy2>