Your message dated Sun, 26 Oct 2014 18:05:58 +0000 with message-id <[email protected]> and subject line Bug#720584: fixed in sysvinit 2.88dsf-57 has caused the Debian Bug report #720584, regarding initscripts.postinst: wrong --compare-versions checks (causing debootstrap fakechroot failure) to be marked as done.
This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact [email protected] immediately.) -- 720584: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=720584 Debian Bug Tracking System Contact [email protected] with problems
--- Begin Message ---Package: initscripts Version: 2.88dsf-43 Severity: important Tags: d-i patch Dear Maintainer, I'm unsuccessfully trying to use debirf to create a Debian testing initrd, i.e. using debootstrap in fakechroot mode. * What led up to the situation? debootstrap fails, with an sh -ex enabled debootstrap.log containing: + info CONFIGURING Configuring %s... initscripts + local name=CONFIGURING + local fmt=Configuring %s... + shift + shift + [ ] + printf I: Configuring %s...\n initscripts + expect=installed + read status pkg qstate sed: couldn't open temporary file /etc/default/sedUlNpID: Permission denied dpkg: error processing initscripts (--configure): subprocess installed post-installation script returned error exit status 4 + [ status: = EXITCODE ] + [ : error : subprocess installed post-installation script returned error exit status 4 = installed ] I then did debirf enter <config> (enter chroot) and there did a manual dpkg -D777 --configure initscripts and then sh -x /var/lib/dpkg/info/initscripts.postinst configure which both successfully exhibited the sed failure above. Narrowing it down some more via a strace -f sed -i 's:foobar:foobaz:' /etc/default/rcS|less showed open("/etc/default/sedO6lzxA", O_RDWR|O_CREAT|O_EXCL|O_LARGEFILE, 0600) = -1 EACCES (Permission denied) umask(022) = 0700 write(2, "sed: ", 5sed: ) = 5 write(2, "couldn't open temporary file /et"..., 70couldn't open temporary file /etc/default/sedO6lzxA: Permission denied) = 70 write(2, "\n", 1 * What exactly did you do (or not do) that was effective (or ineffective)? Nothing. * What was the outcome of this action? Complete debirf (debootstrap) failure so far, for any install of Debian testing. == "WTF" == This failure was not avoided, due to postinst improperly deciding to enter the sed fixup section despite it being an initial-install, thus dpkg --compare-versions with empty $PREV_VER ought to have been false (see dpkg(1)). Or, to put it more clearly, this perhaps unfixable permission failure (see also related bugs #596284 / #694986 / #649148 ) would at least have been avoided in *my* situation of an initial-install - *if* the version checks had been correct. * What outcome did you expect instead? Given that I expect debootstrap to be an absolute hotpath (thousands of server installs) this should have Nothing But Worked. Will soon try non-fakechroot mode of debootstrap instead, since I'm hopeful that this will avoid the Permission Denied issue. I'm quickly running out of ideas on how to get a Debian testing debootstrap up and running, however, since I'm unsure whether I'm able to locally supply a custom-patched package, or use any other suitable methods to avoid that failure. There are several similar Internet reports about debootstrap-related problems in this area. This reportbug template produced by host system (aptosid 2013-01), not semi-bootstrapped one - some info may be slightly(!) mismatching. Untested patch included here. It corrects the comparison operators where suitable, and only there! (some others further below talk of those being necessary for both specific version ranges and initial-install empty-string, too) With this patch it means that postinst now depends on solid availability of the possibly too modern lt-nl variants (perhaps that kind of dpkg support may happen to be not old enough?). Thank you for maintaining a very important and thus sizeable core package! Andreas Mohr --- initscripts.postinst.orig 2013-08-23 15:39:09.021852346 +0200 +++ initscripts.postinst 2013-08-23 15:41:39.228899560 +0200 @@ -61,16 +61,16 @@ } # In 2.88dsf-23 the "mountoverflowtmp" script was dropped entirely. -if dpkg --compare-versions "$PREV_VER" lt "2.88dsf-23" ; then +if dpkg --compare-versions "$PREV_VER" lt-nl "2.88dsf-23" ; then update-rc.d -f mountoverflowtmp remove >/dev/null fi # In 2.88dsf-41+jessie1 the "mtab.sh" script was dropped entirely. -if dpkg --compare-versions "$PREV_VER" lt "2.88dsf-41+jessie1" ; then +if dpkg --compare-versions "$PREV_VER" lt-nl "2.88dsf-41+jessie1" ; then update-rc.d -f mtab.sh remove >/dev/null fi # Comment out obsolete options in rcS. -if dpkg --compare-versions "$PREV_VER" lt "2.88dsf-23" ; then +if dpkg --compare-versions "$PREV_VER" lt-nl "2.88dsf-23" ; then if [ -f /etc/default/rcS ]; then sed -i \ -e 's:^\(RAMRUN=.*\)$:#\1 # OBSOLETE; see /etc/default/tmpfs and tmpfs(5).:' \ -- System Information: Debian Release: 7.0 APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i586) Kernel: Linux 3.9-0.slh.4-aptosid-686 (SMP w/1 CPU core; PREEMPT) Locale: LANG=de_DE.utf8, LC_CTYPE=de_DE.utf8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages initscripts depends on: ii coreutils 8.13-3.3 ii debianutils 4.3.4 ii libc6 2.17-92 ii lsb-base 4.1+Debian9 ii mount 2.20.1-5.3 ii sysv-rc 2.88dsf-41 ii sysvinit-utils 2.88dsf-41 Versions of packages initscripts recommends: ii e2fsprogs 1.42.5-1.1 ii psmisc 22.20-1 initscripts suggests no packages. -- no debconf information
--- End Message ---
--- Begin Message ---Source: sysvinit Source-Version: 2.88dsf-57 We believe that the bug you reported is fixed in the latest version of sysvinit, which is due to be installed in the Debian FTP archive. A summary of the changes between this version and the previous one is attached. Thank you for reporting the bug, which will now be closed. If you have further comments please address them to [email protected], and the maintainer will reopen the bug report if appropriate. Debian distribution maintenance software pp. Petter Reinholdtsen <[email protected]> (supplier of updated sysvinit package) (This message was generated automatically at their request; if you believe that there is a problem with it please contact the archive administrators by mailing [email protected]) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Format: 1.8 Date: Sun, 26 Oct 2014 18:36:37 +0100 Source: sysvinit Binary: sysvinit sysvinit-core sysvinit-utils sysv-rc initscripts bootlogd Architecture: source amd64 all Version: 2.88dsf-57 Distribution: unstable Urgency: medium Maintainer: Debian sysvinit maintainers <[email protected]> Changed-By: Petter Reinholdtsen <[email protected]> Description: bootlogd - daemon to log boot messages initscripts - scripts for initializing and shutting down the system sysv-rc - System-V-like runlevel change mechanism sysvinit - System-V-like init utilities - transitional package sysvinit-core - System-V-like init utilities sysvinit-utils - System-V-like utilities Closes: 587954 606875 641669 720584 742615 745260 745503 745505 746587 751589 Changes: sysvinit (2.88dsf-57) unstable; urgency=low . * Upload to unstable. . sysvinit (2.88dsf-56) experimental; urgency=medium . [ Dimitri John Ledkov ] * service & invoke-rc.d: in upstart interfacing code, check that the job is actually known to upstart. This is because during upgrades, pid 1 might still be an older upstart which may not yet support syntax of the newly unpacked jobs, thus sysv-init script should be continued to be used instead. (Closes: #745503) * service & invoke-rc.d: unset UPSTART_SESSION environment variable to make sure all upstart initctl commands are executed against system init and not the session one. (Closes: #745505) * service: in upstart interfacing code, map "force-reload" to restart as per Debian policy 9.3.2, since there is no way to know for-sure if a reload is supported. . [ Petter Reinholdtsen ] * Adjust the sysvinit/hurd-fix-inittab debconf template to no longer claim sysvinit have never been used on hurd. It is not true any more. . [ Steve Langasek ] * Add necessary Conflicts/Replaces against systemd-sysv and upstart to sysvinit-core. Closes: #751589. * Confirm SRU fixing circular dependency of mountnfs ifupdown hook when running with systemd. Closes: #746587. . [ Petter Reinholdtsen ] * Update Vcs links in control file to the correct paths. * Correct code in update-rc.d looking for insserv to look in /usr/lib/insserv/insserv every time insserv is used, not only only one of the relevant code blocks. . sysvinit (2.88dsf-55.3) experimental; urgency=medium . * Non-maintainer upload. * Since the new "init" metapackage has taken over the role to ensure an init system is installed at all times, drop the Essential: yes flag from sysvinit and demote its priority to optional so this package is no longer pulled in on new installations on Linux. Make sysvinit depend on "init" so this new package is installed on upgrades. * Provide a fallback SysV init binary in the sysvinit package which can be used to boot the system via init=/lib/sysvinit/init even if systemd is the default init system and /sbin/init is provided by systemd-sysv. * Demote the priority of sysvinit-core to extra so it is no longer installed by default on Linux. . sysvinit (2.88dsf-55.2) experimental; urgency=medium . * Fix a formatting error in mountnfs which turned a tab character into U+21A6. . sysvinit (2.88dsf-55.1) experimental; urgency=medium . * Non-maintainer upload. * Skip the mountnfs hook when being triggered by the networking SysV init script and instead use the systemd built-in mechanisms to mount remote file systems. This avoids a deadlock caused by the rpcbind SysV init script depending on $network and the $network LSB facility being provided by the networking SysV init script. (Closes: #746587) . sysvinit (2.88dsf-55) experimental; urgency=medium . [ Gabriele Giacone ] * sysv-rc: - On hurd, fix hurd-console addition to inittab if inittab is already existent and fix getty pathnames in commented out lines as well (Closes: #745260). . [ Petter Reinholdtsen ] * Drop sysv-rc-conf as suggests. It do not work with dependency based boot ordering. * Adjust initscripts.postinst to use --compare-versions checks that cause relevant code to only run on upgrades, and not on first time install, fixing debootstrap fakechroot failure (Closes: #720584). Patch from Andreas Mohr. * Add 'status' support to all init.d scripts (Closes: #641669). Based on patch from Peter Eisentraut. . sysvinit (2.88dsf-54) experimental; urgency=medium . [ Justus Winter ] * hurd-i386 runsystem.sysv: honor init=something in the kernel command line (Closes: #742615). . [ Petter Reinholdtsen ] * Look for insserv /usr/lib/insserv/insserv and /sbin/insserv, to handle newer insserv packages where it no longer is in the default PATH. * Let the service script pass environment variables LANGUAGE, LC_CTYPE, LC_NUMERIC, LC_TIME, LC_COLLATE, LC_MONETARY, LC_MESSAGES, LC_PAPER, LC_NAME, LC_ADDRESS, LC_TELEPHONE, LC_MEASUREMENT, LC_IDENTIFICATION and LC_ALL (in other word, all locale related environment variables) in addition to LANG to the init.d scripts, in accordance with the principle of least surprise (Closes: #606875). * Change checkfs/checkroot and avoid printing the kernel command line during boot if forcefsck is set (Closes: #587954). Checksums-Sha1: 1a595cd557bf8bec85c582623ebe33a7e9f8059f 1819 sysvinit_2.88dsf-57.dsc f7d5eed6d58ab09ded29b2e63f16557750bec78b 152524 sysvinit_2.88dsf-57.debian.tar.xz 2f87ec9bee1ecf0b147427a376c5b839b559747f 99920 sysvinit_2.88dsf-57_amd64.deb 6a43bd2e31067711eb2cbcea37988f5837f3904a 131870 sysvinit-core_2.88dsf-57_amd64.deb 6f6399810f9a2ac98b528e79e89da0aecd6e80cc 84460 sysvinit-utils_2.88dsf-57_amd64.deb 307962a123bb7710affcf6a8eb55c94ee9e5e2f9 82976 sysv-rc_2.88dsf-57_all.deb a38c43a252477057c98d4d3cb6fc37d8a9a24e04 87346 initscripts_2.88dsf-57_amd64.deb 638884fddaac042d7a345152eb5ebf673f219782 58764 bootlogd_2.88dsf-57_amd64.deb Checksums-Sha256: 5cf0120510fc4106ce7287dd6712aa9bdb886129e0be29d8f6e0f6265a81b98b 1819 sysvinit_2.88dsf-57.dsc 5d5f80650220162eafbaab0c115b1edc2bdb55c9a007e11fac546e4ace490740 152524 sysvinit_2.88dsf-57.debian.tar.xz 07704208cb376cba334b2c44fe7ce9d188364b0964400b580474768b4945e10d 99920 sysvinit_2.88dsf-57_amd64.deb da95d2aed4e7419c3fa5c6338c58c808169e12f9bb4ee6ca6babc6082dd51638 131870 sysvinit-core_2.88dsf-57_amd64.deb b80ac7fe195b38dab5f1e077f232bcc8e31fdfa5e4d3613854ddb6c4283d87bc 84460 sysvinit-utils_2.88dsf-57_amd64.deb 7b1bcecde7e800ec223d3bd0da3530fe566105e2aadb5453a9a30cb773107214 82976 sysv-rc_2.88dsf-57_all.deb 8559902ea39ccaaaea0a420c7115cb2cfec3abcfe5f7b4aabf2fac925c0639ae 87346 initscripts_2.88dsf-57_amd64.deb 0fb48edab1a406d9340a767337ba06b4825bdea42151a3d1be7c38f4e423ca7c 58764 bootlogd_2.88dsf-57_amd64.deb Files: 31a5593fdc4d1f787cd120f9ffb690ec 1819 admin required sysvinit_2.88dsf-57.dsc 591ffc01da46a564365aaad0ca898099 152524 admin required sysvinit_2.88dsf-57.debian.tar.xz 291b9c399ca5c8204f35c441896402a0 99920 admin optional sysvinit_2.88dsf-57_amd64.deb d09ed380ed6b2f5face73deff534a32b 131870 admin extra sysvinit-core_2.88dsf-57_amd64.deb 9772eac4b9ea1b42879e4555a01b854b 84460 admin required sysvinit-utils_2.88dsf-57_amd64.deb d80d82081c05b8890fb54c224be197f0 82976 admin required sysv-rc_2.88dsf-57_all.deb 7842bae75757c922d6fedab5c6e071a0 87346 admin required initscripts_2.88dsf-57_amd64.deb 573557811b4ff30f31fb13f7d39f7e1a 58764 admin optional bootlogd_2.88dsf-57_amd64.deb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iD8DBQFUTTUb20zMSyow1ykRAuFqAJ9KmgAWyWigfDGw2HvaKjIVoQxDsACeI5EE jJlLH5aTm5USWaaBRSX7XFs= =A3fP -----END PGP SIGNATURE-----
--- End Message ---
_______________________________________________ Pkg-sysvinit-devel mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-sysvinit-devel
