Dead-lock in replaced package scripts
Hi, I am searching for a solution for this scenario: nsd3 (<= 4.0.0), e.g. from 3.2.x postrm script _remove_ "nsd" user nsd3 (>= 4.0.0) is just a dummy package with hard depends on nsd (>= 4.0.0) nsd (>= 4.0.0) creates "nsd" user in postinst and remove "nsd" user in postrm If nsd3 package is upgraded along nsd package then everything is good. Now there's a rare situation where: nsd3 3.2.x has been removed, but not purged (e.g. postrm hasn't been run) nsd 4.0.0 is installed nsd3 3.2.x is purged Now the "nsd" user has been removed in nsd3.postinst, but it is still needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I just cross the fingers and hope nobody would do such thing? O. -- Ondřej Surý Knot DNS (https://www.knot-dns.cz/) – a high-performance DNS server -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1384258570.14536.46360853.0cc1b...@webmail.messagingengine.com
Re: Dead-lock in replaced package scripts
On Tue, 12 Nov 2013 13:16:10 +0100 Ondřej Surý wrote: > Hi, > > I am searching for a solution for this scenario: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=621833 Generally, I avoid even trying to remove system users during package removal. Remove the files for that user but not the user itself. (What harm does it do to leave the user defined?) > Now the "nsd" user has been removed in nsd3.postinst, but it is still > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or > should I just cross the fingers and hope nobody would do such thing? The old postinst sounds broken. Does the user removal succeed? If the user is currently active, it should fail - which could give you a way to prevent removal, assuming that the old postinst will handle that cleanly. Expected user intervention for this would be to reinstall the newer version of the package - the maintainer scrips should be able to cope in this situation by putting the system user back if the old package removed it. The application may need to also complain loudly if the required user does not exist, so that someone can tell what needs to be done to fix it. -- Neil Williams = http://www.linux.codehelp.co.uk/ signature.asc Description: PGP signature
Re: Dead-lock in replaced package scripts
On Tue, 12 Nov 2013, Ondřej Surý wrote: > Now the "nsd" user has been removed in nsd3.postinst, but it is still > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I > just cross the fingers and hope nobody would do such thing? You probably shouldn't ever remove users. Cheers, -- | .''`. ** Debian ** Peter Palfrader | : :' : The universal http://www.palfrader.org/ | `. `' Operating System | `-http://www.debian.org/ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131112125556.gp...@anguilla.noreply.org
Re: Dead-lock in replaced package scripts
On Tue, Nov 12, 2013, at 13:55, Peter Palfrader wrote: > On Tue, 12 Nov 2013, Ondřej Surý wrote: > > > Now the "nsd" user has been removed in nsd3.postinst, but it is still > > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I > > just cross the fingers and hope nobody would do such thing? > > You probably shouldn't ever remove users. That's very nice answer, but not very helpful, since it doesn't address my question. O. -- Ondřej Surý Knot DNS (https://www.knot-dns.cz/) – a high-performance DNS server -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1384262553.8893.46386409.7574c...@webmail.messagingengine.com
Re: Dead-lock in replaced package scripts
Hi, On Tue, 12 Nov 2013, Ondřej Surý wrote: > On Tue, Nov 12, 2013, at 13:55, Peter Palfrader wrote: > > On Tue, 12 Nov 2013, Ondřej Surý wrote: > > > > > Now the "nsd" user has been removed in nsd3.postinst, but it is still > > > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I > > > just cross the fingers and hope nobody would do such thing? > > > > You probably shouldn't ever remove users. > > That's very nice answer, but not very helpful, since it doesn't address > my question. You don't have many options. You can check the status of the nsd3 package when you install/upgrade nsd and emit a warning pointing to some README.Debian that explains how to clean up the situation (or you can do it automatically, it's not very clean but should be rather straightforward). Or you can add a dependency on nsd3 (>= 4) to nsd for the time of a release cycle, thus forcing the user to keep a dummy package around. In any case, I would still fix the nsd package to not remove the user on postrm... Cheers, -- Raphaël Hertzog ◈ Debian Developer Discover the Debian Administrator's Handbook: → http://debian-handbook.info/get/ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131112133723.gb27...@x230-buxy.home.ouaza.com
Re: Dead-lock in replaced package scripts
]] Ondřej Surý > Now the "nsd" user has been removed in nsd3.postinst, but it is still > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I > just cross the fingers and hope nobody would do such thing? As others have said: stop removing the user, but also, rm the old postrm script (with suitable checks). It's not pretty, but it's sometimes necessary. -- Tollef Fog Heen UNIX is user friendly, it's just picky about who its friends are -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87fvr1aghc@qurzaw.varnish-software.com
Re: Dead-lock in replaced package scripts
On Tue, Nov 12, 2013 at 12:52:35PM +, Neil Williams wrote: > Generally, I avoid even trying to remove system users during package > removal. Remove the files for that user but not the user itself. (What > harm does it do to leave the user defined?) This seems very sensible and pragmatic, but your parenthetical reminds me of Russ's recent thread here about system accounts with login shells. Perhaps in situations like these switching the shell to /bin/false might not be a bad compromise. -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131112144631.ga19...@bryant.redmars.org
Re: Dead-lock in replaced package scripts
On Tue, Nov 12, 2013, at 14:37, Tollef Fog Heen wrote: > ]] Ondřej Surý > > > Now the "nsd" user has been removed in nsd3.postinst, but it is still > > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I > > just cross the fingers and hope nobody would do such thing? > > As others have said: stop removing the user I will, but the harm has been already done at the time I become the nsd* maintainer. > but also, rm the old postrm script (with suitable checks). I'll use: sed -i -e '/deluser --quiet nsd/ d' /var/lib/dpkg/info/nsd3.postrm instead, but thanks for the hint. That helped me to break from '/var/lib/dpkg/info is sacred' mindset. > It's not pretty, but it's sometimes necessary. On Tue, Nov 12, 2013, at 14:37, Raphael Hertzog wrote: > You can check the status of the nsd3 package when you install/upgrade nsd > and emit a warning pointing to some README.Debian that explains how to > clean up the situation (or you can do it automatically, it's not very > clean but should be rather straightforward). > > Or you can add a dependency on nsd3 (>= 4) to nsd for the time of a > release cycle, thus forcing the user to keep a dummy package around. That would create a dependency loop (since nsd3 needs to depend on nsd (>= 4)). > In any case, I would still fix the nsd package to not remove the user on > postrm... Already done in 4.0.0-3 (that is almost finished and waiting for a solution for this). Thanks to you both, this helped me to find the correct solution to remedy the problem. Ondrej -- Ondřej Surý Knot DNS (https://www.knot-dns.cz/) – a high-performance DNS server -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1384273948.22526.46487449.337a2...@webmail.messagingengine.com
Bug#729432: ITP: php-sabre-event -- lightweight library for event management
Package: wnpp Severity: wishlist Owner: David Prévot * Package name: php-sabre-event Version : 1.0.0 Upstream Author : Evert Pot * URL : https://github.com/evert/sabre-vobject * License : BSD-3-clause Programming Lang: PHP Description : lightweight library for event management sabre/event’s design is inspired by Node.js's EventEmitter. . It provides some features like prioritization, and the ability to stop the event chain, like JavaScript’s preventDefault. It’s a dependency of the upcoming php-sabredav 1.9 branch that will replace the current 1.8 branch in experimental (the 1.7 branch will stay in unstable as long as owncloud depends on it). I intend to handle its packaging inside the PHP PEAR (and Composer) team. Regards David signature.asc Description: Digital signature
Bug#729433: ITP: php-sabre-http -- library to ease the work with HTTP protocol
Package: wnpp Severity: wishlist Owner: David Prévot Control: block -1 by 729432 * Package name: php-sabre-http Version : 2.0.0~alpha5 Upstream Author : Evert Pot * URL : https://github.com/evert/sabre-vobject * License : BSD-3-clause Programming Lang: PHP Description : library to ease the work with HTTP protocol sabre/http provides a toolkit to ease the work with HTTP protocol. . This library is a wrapper around the following PHP constructs: . For Input: * $_GET * $_POST * $_SERVER * php://input or $HTTP_RAW_POST_DATA. For output: * php://output or echo. * header() . It provides a Request object, and a Response object. The objects are extensible and easily mockable. It’s a dependency of the upcoming php-sabredav 1.9 branch that will replace the current 1.8 branch in experimental (the 1.7 branch will stay in unstable as long as owncloud depends on it). I intend to handle its packaging inside the PHP PEAR (and Composer) team. Regards David signature.asc Description: Digital signature
Re: Dead-lock in replaced package scripts
On Tue, Nov 12, 2013 at 05:32:28PM +0100, Ondřej Surý wrote: > On Tue, Nov 12, 2013, at 14:37, Tollef Fog Heen wrote: > > ]] Ondřej Surý > > > > > Now the "nsd" user has been removed in nsd3.postinst, but it is still > > > needed by nsd (>= 4.0.0) package. Any ideas how to fix that? Or should I > > > just cross the fingers and hope nobody would do such thing? > > > > As others have said: stop removing the user > > I will, but the harm has been already done at the time I become the nsd* > maintainer. > > > but also, rm the old postrm script (with suitable checks). > > I'll use: > > sed -i -e '/deluser --quiet nsd/ d' /var/lib/dpkg/info/nsd3.postrm > > instead, but thanks for the hint. That helped me to break from > '/var/lib/dpkg/info is sacred' mindset. It's the hard-coding of the path that should be avoided. Instead, use “dpkg-query --control-path nsd3 postrm” to get the path. Cheers, -- James GPG Key: 4096R/331BA3DB 2011-12-05 James McCoy signature.asc Description: Digital signature
Bug#729441: ITA: newlib -- Newlib is a C library intended for use on embedded systems
Package: wnpp Severity: wishlist Owner: Agustin Henze X-Debbugs-CC: debian-devel@lists.debian.org, debian-cr...@lists.debian.org Hi all, I'm adopting this package and I'll upload the new upstream release (2.0.0) with big changes in the packaging. The main change is that it'll only generate 2 binaries. * libnewlib-dev: This package contains the newlib headers. * libnewlib-arm-none-eabi: This package contains the newlib library compiled for Cortex-A*, Cortex-R4/R5/R7 and Cortex-M0/M0+/M3/M4 targets. That's because I use and can test daily those targets. If someone can help for still maintain the old binaries packages (libnewlib0, newlib-spu, newlib-m68hc1x, newlib-source) is welcome. I'll working in collab-maint[0] when alioth back online again. [0] git://anonscm.debian.org/collab-maint/newlib.git -- TiN signature.asc Description: OpenPGP digital signature
