February 2012 - Arch-projects - lists.archlinux.org

[arch-projects] netcfg: fixes and features
by Jouke Witteveen 21 Mar '12

21 Mar '12

Dear Thomas and the rest, A few of us have been looking at the netcfg code lately. One thing that came to light was that FS#24599 is actualy a wpa_actiond bug, so, Thomas, could you take it upon you? Furthermore, Remy's development tree held 8 bugfixes and some improvements, which I augmented with another 5.5 bugfixes and 3.5 added features (the half one being a syntax to reconnect based on an interface, as a fix for FS#28196). I would like to see someone package this work, that is available at https://github.com/joukewitteveen/netcfg and push a package to the (testing?) repository. I am not requesting maintainership of the code (Alfredo Palhares is after it too), but I am willing to be the maintainer. My request is just to move netcfg forward. Regards, - Jouke List of fixed bugs outside Remy's tree: FS#23934 FS#26370 FS#26607 FS#27496 (not quite) FS#28022 FS#28196

7 20

[arch-projects] [PATCH 1/4] fhs: remove /var/lock form valid paths
by Tom Gundersen 28 Feb '12

28 Feb '12

It is already forbidden, so should not be valid. Signed-off-by: Tom Gundersen <teg(a)jklm.no> --- Namcap/rules/fhs.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Namcap/rules/fhs.py b/Namcap/rules/fhs.py index 2ce34d0..28cb30f 100644 --- a/Namcap/rules/fhs.py +++ b/Namcap/rules/fhs.py @@ -29,7 +29,7 @@ class FHSRule(TarballRule): 'bin/', 'etc/', 'lib/', 'sbin/', 'opt/', 'srv/', 'usr/bin/', 'usr/include/', 'usr/lib/', 'usr/lib32/', 'usr/sbin/', 'usr/share/', - 'var/cache/', 'var/lib/', 'var/lock/', 'var/log/', 'var/opt/', + 'var/cache/', 'var/lib/', 'var/log/', 'var/opt/', 'var/spool/', 'var/state/', '.PKGINFO', '.INSTALL', '.CHANGELOG', ] -- 1.7.9.2

7 15

[arch-projects] [PATCH] Add --no-user-install to PKGBUILD-rubygem.proto
by Peter Lewis 27 Feb '12

27 Feb '12

This makes the prototype compatible with the recent ruby changes. FS #28681 Signed-off-by: Peter Lewis <plewis(a)aur.archlinux.org> --- prototypes/PKGBUILD-rubygem.proto | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/prototypes/PKGBUILD-rubygem.proto b/prototypes/PKGBUILD-rubygem.proto index 8adfff9..84fbd70 100644 --- a/prototypes/PKGBUILD-rubygem.proto +++ b/prototypes/PKGBUILD-rubygem.proto @@ -27,7 +27,7 @@ package() { # the system, makepkg will exit with an error when sourcing the PKGBUILD. local _gemdir="$(ruby -rubygems -e'puts Gem.default_dir')" - gem install --ignore-dependencies -i "$pkgdir$_gemdir" -n "$pkgdir/usr/bin" \ + gem install --no-user-install --ignore-dependencies -i "$pkgdir$_gemdir" -n "$pkgdir/usr/bin" \ "$_gemname-$pkgver.gem" } -- 1.7.9.2

1 1

[arch-projects] [RFC] [PATCH] [mkarchroot] Add -d param to optionally support devtmpfs
by Gerardo Exequiel Pozzi 26 Feb '12

26 Feb '12

The main intention of this patch is to take advantage of /dev/loop-control and loop devices. Signed-off-by: Gerardo Exequiel Pozzi <vmlinuz386(a)yahoo.com.ar> --- mkarchroot.in | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-) diff --git a/mkarchroot.in b/mkarchroot.in index cda761c..d45cfba 100644 --- a/mkarchroot.in +++ b/mkarchroot.in @@ -13,6 +13,7 @@ m4_include(lib/common.sh) FORCE='n' RUN='' NOCOPY='n' +USE_DEVTMPFS='n' working_dir='' @@ -29,11 +30,12 @@ usage() { echo ' -M <file> Location of a makepkg config file' echo ' -n Do not copy config files into the chroot' echo ' -c <dir> Set pacman cache. Default: /var/cache/pacman/pkg' + echo ' -d Use devtmpfs instead of minimal /dev (tmpfs)' echo ' -h This message' exit 1 } -while getopts 'r:ufnhC:M:c:' arg; do +while getopts 'r:ufnhC:M:c:d' arg; do case "${arg}" in r) RUN="$OPTARG" ;; u) RUN='/bin/sh -c "pacman -Syu --noconfirm && (pacman -Qqu >/dev/null && pacman -Su --noconfirm || exit 0)"' ;; @@ -42,6 +44,7 @@ while getopts 'r:ufnhC:M:c:' arg; do M) makepkg_conf="$OPTARG" ;; n) NOCOPY='y' ;; c) cache_dir="$OPTARG" ;; + d) USE_DEVTMPFS="y" ;; h|?) usage 0 ;; *) error "invalid argument '${arg}'"; usage ;; esac @@ -90,20 +93,24 @@ chroot_mount() { mount -t proc proc "${working_dir}/proc" [[ -e "${working_dir}/dev" ]] || mkdir "${working_dir}/dev" - mount -t tmpfs dev "${working_dir}/dev" -o mode=0755,size=10M,nosuid - mknod -m 666 "${working_dir}/dev/null" c 1 3 - mknod -m 666 "${working_dir}/dev/zero" c 1 5 - mknod -m 600 "${working_dir}/dev/console" c 5 1 - mknod -m 644 "${working_dir}/dev/random" c 1 8 - mknod -m 644 "${working_dir}/dev/urandom" c 1 9 - mknod -m 666 "${working_dir}/dev/tty" c 5 0 - mknod -m 666 "${working_dir}/dev/tty0" c 4 0 - mknod -m 666 "${working_dir}/dev/full" c 1 7 - ln -s /proc/kcore "${working_dir}/dev/core" - ln -s /proc/self/fd "${working_dir}/dev/fd" - ln -s /proc/self/fd/0 "${working_dir}/dev/stdin" - ln -s /proc/self/fd/1 "${working_dir}/dev/stdout" - ln -s /proc/self/fd/2 "${working_dir}/dev/stderr" + if [[ ${USE_DEVTMPFS} = "y" ]]; then + mount -t devtmpfs dev "${working_dir}/dev" -o mode=0755,nosuid + else + mount -t tmpfs dev "${working_dir}/dev" -o mode=0755,size=10M,nosuid + mknod -m 666 "${working_dir}/dev/null" c 1 3 + mknod -m 666 "${working_dir}/dev/zero" c 1 5 + mknod -m 600 "${working_dir}/dev/console" c 5 1 + mknod -m 644 "${working_dir}/dev/random" c 1 8 + mknod -m 644 "${working_dir}/dev/urandom" c 1 9 + mknod -m 666 "${working_dir}/dev/tty" c 5 0 + mknod -m 666 "${working_dir}/dev/tty0" c 4 0 + mknod -m 666 "${working_dir}/dev/full" c 1 7 + ln -s /proc/kcore "${working_dir}/dev/core" + ln -s /proc/self/fd "${working_dir}/dev/fd" + ln -s /proc/self/fd/0 "${working_dir}/dev/stdin" + ln -s /proc/self/fd/1 "${working_dir}/dev/stdout" + ln -s /proc/self/fd/2 "${working_dir}/dev/stderr" + fi [[ -e "${working_dir}/dev/shm" ]] || mkdir "${working_dir}/dev/shm" mount -t tmpfs shm "${working_dir}/dev/shm" -o nodev,nosuid,size=128M -- 1.7.9.2

1 0

[arch-projects] [PATCH][namcap] use pure python for exec stack check
by Dave Reisner 26 Feb '12

26 Feb '12

Instead of pulling prelink as a dependency, use a pure python library, pyelftools, for reading the program headers. This could be useful elsewhere to replace parsing of things such as readelf. --- This relies on a pure python ELF parsing lib I found, pyelftools, to do the job of prelink's execstack. I sort of pieced this together from example code in pyelftools, so I'm not sure it's all that wonderful, but imo it's better than relying on prelink it does seem to get the job done. Namcap/rules/elffiles.py | 23 +++++++++++++++-------- 1 files changed, 15 insertions(+), 8 deletions(-) diff --git a/Namcap/rules/elffiles.py b/Namcap/rules/elffiles.py index 135b6d5..f8f16ac 100644 --- a/Namcap/rules/elffiles.py +++ b/Namcap/rules/elffiles.py @@ -22,6 +22,8 @@ import os import tempfile import subprocess +from elftools.elf.elffile import ELFFile + from Namcap.util import is_elf, clean_filename from Namcap.ruleclass import * @@ -107,8 +109,9 @@ class ELFExecStackRule(TarballRule): """ Check for executable stacks in ELF files. - Introduced by FS#26458. This uses the execstack utility from - the prelink package. + Introduced by FS#26458. Uses pyelftools to read the GNU_STACK + program header and ensure it does not have the executable bit + set. """ name = "elfexecstack" @@ -123,12 +126,16 @@ class ELFExecStackRule(TarballRule): continue try: - proc = subprocess.Popen(["execstack", tmpname], - stdout=subprocess.PIPE, - stderr=subprocess.PIPE) - out, err = proc.communicate() - if out.startswith(b'X'): - exec_stacks.append(entry.name) + fp = open(tmpname, 'rb') + elffile = ELFFile(fp) + + for segment in elffile.iter_segments(): + if segment['p_type'] != 'PT_GNU_STACK': continue + + mode = segment['p_flags'] + if mode & 1: exec_stacks.append(entry.name) + + fp.close() finally: os.unlink(tmpname) -- 1.7.9

2 1

[arch-projects] [PATCH] namcap: Add rubypaths rule warning against usage of usr/lib/ruby/site_ruby.
by Rémy Oudompheng 26 Feb '12

26 Feb '12

--- This patch implements FS#28478. Can you take a look? NEWS | 6 ++++++ Namcap/rules/fhs.py | 9 +++++++++ namcap-tags | 1 + 3 files changed, 16 insertions(+) diff --git a/NEWS b/NEWS index 3ffe3b6..cbc166d 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,9 @@ +namcap (3.2.3) + - add pyelftools as a dependency, drop prelink. + - add rubypaths rule warning against use of site_ruby + - modify elfexecstack rule to use elftools + - add test case for elfexecstack + namcap (3.2.2) - fix: unwanted warning when testing database is not found (FS#27973) - fix: Creative Commons licenses not recognized (FS#27489) diff --git a/Namcap/rules/fhs.py b/Namcap/rules/fhs.py index 2ce34d0..e66d67e 100644 --- a/Namcap/rules/fhs.py +++ b/Namcap/rules/fhs.py @@ -90,4 +90,13 @@ class FHSInfoPagesRule(TarballRule): if part == "info": self.warnings.append(("potential-non-fhs-info-page %s", i.name)) +class RubyPathsRule(TarballRule): + name = "rubypaths" + description = "Verifies correct usage of folders by ruby packages" + def analyze(self, pkginfo, tar): + for i in tar.getmembers(): + if i.name.startswith('usr/lib/ruby/site_ruby'): + self.warnings.append(("site-ruby", ())) + return + # vim: set ts=4 sw=4 noet: diff --git a/namcap-tags b/namcap-tags index fce8cf4..818c7a5 100644 --- a/namcap-tags +++ b/namcap-tags @@ -68,6 +68,7 @@ potential-non-fhs-info-page %s :: Potential non-FHS info page (%s) found. potential-non-fhs-man-page %s :: Potential non-FHS man page (%s) found. script-link-detected %s in %s :: Script link detected (%s) in file %s scrollkeeper-dir-exists %s :: Scrollkeeper directory exists (%s). Remember to not run scrollkeeper till post_{install,upgrade,remove}. +site-ruby :: Found usr/lib/ruby/site_ruby in package, usr/lib/ruby/vendor_ruby should be used instead. specific-host-type-used %s :: Reference to %s should be changed to $CARCH specific-sourceforge-mirror :: Attempting to use specific sourceforge mirror, use downloads.sourceforge.net instead symlink-found %s points to %s :: Symlink (%s) found that points to %s -- 1.7.9.2

1 0

[arch-projects] [initscripts] [GIT] Arch Linux initscripts repository branch master updated. 2012.02.2-1-g7f42e7e
by tomegun＠archlinux.org 24 Feb '12

24 Feb '12

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "Arch Linux initscripts repository". The branch, master has been updated via 7f42e7e621369aa6eaab446dc4c1bc3f04c4b266 (commit) from fd54cc63cd6853b839d8a0d15ff7c8506743ac5a (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit 7f42e7e621369aa6eaab446dc4c1bc3f04c4b266 Author: Tom Gundersen <teg(a)jklm.no> Date: Fri Feb 24 22:36:33 2012 +0100 Revert "deal with both valid return codes of mount -a" This reverts commit fd54cc63cd6853b839d8a0d15ff7c8506743ac5a. The correct return codes of "mount -a" are: 0 all mounts succeeded 64 some mounts succeeded and some failed 32 all mounts failed We want to show "FAILED" on all return codes except for 0. There is a bug in the new mount where "ignored" is taken to be "failed", which gives some false negative. This will be fixed in util-linux. ----------------------------------------------------------------------- Summary of changes: functions | 1 - 1 files changed, 0 insertions(+), 1 deletions(-) hooks/post-receive -- Arch Linux initscripts repository

2 1

Re: [arch-projects] [initscripts] needs update (WAS: [arch-dev-public] new mount in testing)
by Gerardo Exequiel Pozzi 24 Feb '12

24 Feb '12

On 02/24/2012 02:18 PM, Thomas Bächler wrote: > Am 24.02.2012 18:13, schrieb Tom Gundersen: >> With util-linux-2.21 we have switched to a new implementation of >> mount, based on the libmount library. We have tested this in the >> release candidates for some weeks, and it seems good to go. > I would like to point out that upstream considers this "experimental" at > this point. > >> However, I intend to keep util-linux in testing longer than normal >> just in case. Please keep an eye out for oddities, especially if you >> are using mount in any non-standard ways that we might not have >> tested. > Gerardo will not be happy about this delay (he needs the new losetup). > What about pushing 2.21-1 with "old mount" to core quickly and then > putting 2.21-2 with "new mount" into testing? > I can wait, no problem for me. Anyway I remember, that Dave says something about "new mount" will be disabled for 2.21, and enabled in 2.21.1. I can remember that this "new mount" causes "-a" works in different way than "classic mount" (but is an expected behaviour), and initscripts should be updated to reflect it. (its shows a non-harm FAIL) [#1] http://comments.gmane.org/gmane.linux.utilities.util-linux-ng/5228 [mount (new): Using -a, exits with 64] -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1

2 1

[arch-projects] [initscripts] [GIT] Arch Linux initscripts repository branch master updated. 2012.02.2
by tomegun＠archlinux.org 24 Feb '12

24 Feb '12

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "Arch Linux initscripts repository". The branch, master has been updated via fd54cc63cd6853b839d8a0d15ff7c8506743ac5a (commit) via 248c6f1679343f9e75647bf3b2fefc53c12ba058 (commit) from 6f171422cbb1c48da3f1f440148e163af2e2e635 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit fd54cc63cd6853b839d8a0d15ff7c8506743ac5a Author: Tom Gundersen <teg(a)jklm.no> Date: Fri Feb 24 21:38:26 2012 +0100 deal with both valid return codes of mount -a Signed-off-by: Tom Gundersen <teg(a)jklm.no> commit 248c6f1679343f9e75647bf3b2fefc53c12ba058 Author: Jonathan Frazier <eyeswide(a)gmail.com> Date: Mon Feb 13 03:48:54 2012 -0500 Update the rc.conf manpage for kmod kmod has a modprobe.d man page not modprobe.conf Signed-off-by: Jonathan Frazier <eyeswide(a)gmail.com> ----------------------------------------------------------------------- Summary of changes: functions | 1 + rc.conf.5.txt | 2 +- 2 files changed, 2 insertions(+), 1 deletions(-) hooks/post-receive -- Arch Linux initscripts repository

1 0

[arch-projects] [initscripts] [GIT] Arch Linux initscripts repository annotated tag 2012.02.2 created. 2012.02.2
by tomegun＠archlinux.org 24 Feb '12

24 Feb '12

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "Arch Linux initscripts repository". The annotated tag, 2012.02.2 has been created at 9aae1a54c18bed7b187b9fc645b2c1682b471b2b (tag) tagging fd54cc63cd6853b839d8a0d15ff7c8506743ac5a (commit) replaces 2012.02.1 tagged by Tom Gundersen on Fri Feb 24 21:43:57 2012 +0100 - Log ----------------------------------------------------------------- deal with both valid return codes of mount -a Jonathan Frazier (1): Update the rc.conf manpage for kmod Tom Gundersen (1): deal with both valid return codes of mount -a ----------------------------------------------------------------------- hooks/post-receive -- Arch Linux initscripts repository

1 0