Bug#929063: init: delegate selinux operation to separate binary

2019-05-22 Thread Thorsten Glaser
On Wed, 22 May 2019, Laurent Bigonville wrote: > So let's be it clear for the record. I'll personally oppose all patches that > would undermine the consistency and the experience of using SELinux in debian. Erm… all the patch does is move the SELinux call into a separate executable so that init

Bug#929293: e2fsprogs: e2scrub_all: illegal option -- C (from cronjob)

2019-05-20 Thread Thorsten Glaser
Package: e2fsprogs Version: 1.45.1-2 Severity: normal The cronjob already knows it, but the executable run doesn’t: From: Cron Daemon Message-ID: <20190521011002.0272d220...@tglase.lan.tarent.de> To: r...@tglase.lan.tarent.de Date: Tue, 21 May 2019 03:10:01 +0200 (CEST) Subject: Cron test -e

Bug#920373: default soundfonts (was Re: gstreamer1.0-plugins-bad: no midi sound - gstreamer selects wrong soundfont by default)

2019-05-20 Thread Thorsten Glaser
>On Sat, 18 May 2019 20:34:57 +0100 Tim Colgate > >wrote: >> Unfortunately there doesn't seem to be a standard way in Linux of >>setting a default soundfont to be used by all midi players; I was This (#929185) may also affect timidity, which has its own format, but with a trivial config file

Bug#929185: default soundfonts (was Re: gstreamer1.0-plugins-bad: no midi sound - gstreamer selects wrong soundfont by default)

2019-05-20 Thread Thorsten Glaser
Dixi quod… >Hi Fabian and others, >>Debian. If we add these to the alternatives system to provide >Should we also make all packages providing an alternative for this >Provides some virtual package, for others to depend on? I’d suggest >sf2-soundfont and sf3-soundfont for naming, and SF3

Bug#929185: gstreamer1.0-plugins-bad: no midi sound - gstreamer selects wrong soundfont by default

2019-05-20 Thread Thorsten Glaser
Hi Fabian and others, >I believe we have at least three soundfonts in SF2 format packaged in >Debian. If we add these to the alternatives system to provide >/usr/share/soundfonts/default.sf2, the effort would be manageable. I’d be happy to add an /usr/share/sounds/sf2/default.sf2 alternative to

Bug#929248: apt: E: Repository # changed its 'Suite' value from 'buster' to 'testing'; but how to accept?

2019-05-19 Thread Thorsten Glaser
Package: apt Version: 1.8.0 Severity: normal (but ought to be release-critical, see last paragraph) E: Repository 'http://debs.tarent.de buster InRelease' changed its 'Suite' value from 'buster' to 'testing' N: This must be accepted explicitly before updates for this repository can be

Bug#928919: patch: Do not hide errors from initscripts

2019-05-18 Thread Thorsten Glaser
On Sat, 18 May 2019, Dmitry Bogatov wrote: > If initscript is called during postinst and fail -- yes, it interrupts > installation. And it is bug in package. No, I can install a package and not have it start perfectly fine. Worse is breakage on package *removal*, especially in situations where

Bug#888743: pidofproc returns PIDs in foreign chroots and containers

2019-05-17 Thread Thorsten Glaser
Hi *, I’ve got an initscript in which I cannot use pidofproc for various reasons so I use start-stop-daemon (in the dpkg package) for this. Does this also affect that? I do not use lxc, container stuff, etc. so I’ve got neither any idea about them nor any way to test this. (If so I will make a

Bug#928977: e2fsprogs: e2scrub_all cron job complains about lack of lvm2

2019-05-17 Thread Thorsten Glaser
Package: e2fsprogs Version: 1.45.1-1 Followup-For: Bug #928977 The patch from Message #26 looks good, and I would welcome it. (The idea, to exit silently if only the prerequisite is missing, and only if called as cronjob or newspeak equivalent, is good.) -- System Information: Debian Release:

Bug#794466: Virtualbox backport for Stretch?

2019-05-17 Thread Thorsten Glaser
On Fri, 17 May 2019, Harald Dunkel wrote: > AFAIU #794466 seems to be a political issue. Debian wants to provide Nah, it’s an Oracle issue. They did the same with MySQL IIRC, which has nowadays been replaced by MariaDB as this is untenable with the reliability promises Debian gives. > and

Bug#928919: patch: Do not hide errors from initscripts

2019-05-16 Thread Thorsten Glaser
On Thu, 16 May 2019, Dmitry Bogatov wrote: > Having trailing ":" would result of ignoring result these "do_*" functions. This is normally right. Initscripts must return 0 in most cases, and only explicitly exit with a nonzero errorlevel when needed. > Exit status of initscript does not affect

Bug#929063: init: delegate selinux operation to separate binary

2019-05-16 Thread Thorsten Glaser
On Thu, 16 May 2019, Dmitry Bogatov wrote: > + if (fork() == 0) { /* child */ if ((ret = fork()) == 0) { /* child */ > + execl(SELINUX_CHECK, SELINUX_CHECK, NULL); > + fprintf(stderr, "Failed to execute helper to init SELinux\n"); > +

Bug#910079: Re: apt: fails to “download” local files saying it lacks permissions despite having them

2019-05-15 Thread Thorsten Glaser
Santiago Vila dixit: >What Thorsten reported is a real failure. Mine is just an undesired >warning. (Not sure if they have a common cause) Might be. I'm seeing it too. You suggested... >W: Not enough privileges to perform sandboxed download ... but a per-sources.list-line option to turn off

Bug#928919: patch: Do not hide errors from initscripts

2019-05-13 Thread Thorsten Glaser
On Mon, 13 May 2019, Dmitry Bogatov wrote: > @@ -65,5 +65,3 @@ case "$1" in > exit 3 > ;; > esac > - > -: I think this is not advisable. What this will do is cause the last errorlevel to be returned by the script, which may not be an error but simply the result of a test command.

Bug#810384: crda: /lib/crda/setregdomain fails with exit code 2, /sbin/crda with exit code 255

2019-05-08 Thread Thorsten Glaser
I ran into the same issue: May 8 20:31:30 tglase-nb vmunix: [ 13.348678] systemd-udevd[380]: Process '/lib/crda/setregdomain' failed with exit code 2. May 8 20:31:47 tglase-nb vmunix: [ 32.728410] systemd-udevd[2259]: Process '/sbin/crda' failed with exit code 255. I had only seen the

Bug#787259: Acknowledgement (start error with pbuilder-satisfydepends-gdebi)

2019-05-01 Thread Thorsten Glaser
François Poirotte dixit: > this and I'm not sure how this could be fixed (does the Debian policy even > state what the character encoding should be for the control file?) It requires UTF-8 encoding, has been doing so since a couple of years. Can you try rebuilding without your local hack, but

Bug#927862: youtube-dl: please update to 2019.04.24

2019-04-25 Thread Thorsten Glaser
Package: youtube-dl Version: 2019.01.17-1 Followup-For: Bug #927862 Control: severity -1 grave Control: retitle -1 youtube-dl: needs upstream version update to continue working It now doesn’t work at all any more, however, the upstream version does: $ youtube-dl -f 18 U2n5aGqou9E [youtube]

Bug#910696: openjdk-11-jdk-headless:x32: broken-symlink /usr/lib/jvm/java-11-openjdk-x32/src.zip (and one more)

2019-04-25 Thread Thorsten Glaser
Package: openjdk-11-jdk-headless Version: 11.0.3+7-2 Followup-For: Bug #910696 openjdk-11-jdk-headless:x32: broken-symlink /usr/lib/jvm/java-11-openjdk-x32/src.zip -> ../openjdk-11/src.zip tglase@tglase:~ $ ll /usr/lib/jvm/java-11-openjdk-x32/src.zip lrwxrwxrwx 1 root root 21 Apr 18 04:54

Bug#926851: firefox-esr: [GFX1-]: Failed to lock new back buffer.

2019-04-24 Thread Thorsten Glaser
On Thu, 11 Apr 2019, Thorsten Glaser wrote: > This worked before some upgrade. It also works on my laptop, which has amd64 natively, but not in an amd64 chroot on another system (with the very same amd64 kernel). bye, //mirabilos -- tarent solutions GmbH Rochusstraße 2-4, D-53123 Bonn • h

Bug#927296: unblock: musescore/2.3.2+dfsg2-5

2019-04-18 Thread Thorsten Glaser
und for DEP 5 syntax in a complex case + * Fix share/plugins/notenames.qml and general copyright years + * Correct copyright info for the demos + + -- Thorsten Glaser Thu, 18 Apr 2019 14:01:18 +0200 + +musescore (2.3.2+dfsg2-5) unstable; urgency=medium + + * debian/patches/experiments/

Bug#927296: unblock: musescore/2.3.2+dfsg2-5

2019-04-17 Thread Thorsten Glaser
On Wed, 17 Apr 2019, Thorsten Glaser wrote: > it should sit in sid for a bit longer, that’s no problem either. Oops, I apparently fsck’d up (but why this didn’t show up in lintian for -5 but does show up for the upload in the experimental branch I’m currently doing, I’ll never know) in DE

Bug#927296: unblock: musescore/2.3.2+dfsg2-5

2019-04-17 Thread Thorsten Glaser
complicated: + they were originally PD “Written by Jezar at Dreampoint”, + but modified by Werner who, as EU citizen, cannot relinquish + copyright, so they are, effectively, published under the same + terms as the rest of MuseScore (GPLv2); document that + + -- Thorsten Glaser Tue, 16

Bug#923675: Add related bug #916690 info

2019-04-16 Thread Thorsten Glaser
Daniel Lange dixit: > Thorsten Glaser (CC) has produced a prototype early-rng-init-tools (cf. > https://lists.debian.org/debian-devel/2019/02/msg00327.html) which could be > extended to try reading entropy off the network when it doesn't have a > carried-over seed (as in the Debian In

Bug#926856: firefox: crashes after a recent upgrade

2019-04-11 Thread Thorsten Glaser
Package: firefox Version: 66.0.1-1 Severity: important This one doesn’t even display the profile chooser, it just crashes, although the dialogue to send infos to Mozilla comes up. Script started on 2019-04-11 12:36:27+02:00 [TERM="xterm" TTY="/dev/pts/0" COLUMNS="113" LINES="35"]

Bug#926851: firefox-esr: [GFX1-]: Failed to lock new back buffer.

2019-04-11 Thread Thorsten Glaser
Package: firefox-esr Version: 60.6.1esr-1 Severity: important It just spews tons of this line… [GFX1-]: Failed to lock new back buffer. … on the terminal; the “Firefox - Choose User Profile” window is completely black. This worked before some upgrade. -- Package-specific info: -- Addons

Bug#926707: cups: please do not generate files under /etc but use /var/lib for that (etckeeper woes)

2019-04-10 Thread Thorsten Glaser
# mention etckeeper in the title (easier found thus) retitle 549673 cups: constantly modify configuration files in /etc (bad for etckeeper) retitle 926707 cups-browsed: generates PPDs in /etc, not /var/lib (bad for etckeeper) # merging, as these are effectively the same issue unblock 926707 by

Bug#926707: cups: please do not generate files under /etc but use /var/lib for that (etckeeper woes)

2019-04-09 Thread Thorsten Glaser
Package: cups-browsed Version: 1.21.6-4 Severity: normal With etckeeper on a laptop that often is taken around to different networks, I see lines like these on apt-get --purge dist-upgrade all the time: [master 48dee15] saving uncommitted changes in /etc prior to apt run […] create mode 100644

Bug#925473: tomcat9: sysvinit script missing (Policy §9.11¶2 “must”)

2019-04-07 Thread Thorsten Glaser
tomatically anyway (Closes: #925928) + * Correct the ownership and permissions on the log directory: +group adm and setgid (Closes: #925929) + + -- Thorsten Glaser Tue, 02 Apr 2019 22:54:17 +0200 + tomcat9 (9.0.16-3) unstable; urgency=medium * Removed read/write access to /var/lib/sol

Bug#926561: RFP: micropython -- Unix/Linux port of MicroPython for board-independent use on the host

2019-04-06 Thread Thorsten Glaser
Package: wnpp Severity: wishlist * Package name: micropython Version : 1.10 Upstream Author : Damien George and others * URL : https://github.com/micropython/micropython#the-unix-version * License : MIT Programming Lang: C, Assembly, Python Description

Bug#925473: tomcat9: sysvinit script missing (Policy §9.11¶2 “must”)

2019-04-03 Thread Thorsten Glaser
On Wed, 3 Apr 2019, Emmanuel Bourg wrote: > > I really insist on being able to install tomcat9 without having to > > install a whole other init system, even if it is not used. > > See this as a compromise? I don’t know… the missing initscript is an RC bug, so the compromise would start _after_

Bug#925473: tomcat9: sysvinit script missing (Policy §9.11¶2 “must”)

2019-04-03 Thread Thorsten Glaser
On Wed, 3 Apr 2019, Emmanuel Bourg wrote: > Assuming #926316 gets fixed, I think we should focus only on providing a > usable sysvinit script as required by the policy. Supporting people I really insist on being able to install tomcat9 without having to install a whole other init system, even if

Bug#925473: tomcat9: sysvinit script missing (Policy §9.11¶2 “must”)

2019-04-02 Thread Thorsten Glaser
Hi Emmanuel, >What is the issue with the dependency on systemd? Most people using Debian without systemd have APT pinning or other measures in place that prevent the systemd package, which ships the systemd-sysusers binary (and service?), from being installed, in order to not sneakily being

Bug#925473: tomcat9: sysvinit script missing (Policy §9.11¶2 “must”)

2019-04-02 Thread Thorsten Glaser
Hi Emmanuel, > This restores the ability to create the tomcat user without systemd. due to your objection against perceived complexity, I changed the way I’ve implemented this. Doing this at all is required because the hard “Depends: systemd” will not work on many non-systemd systems, and, as

Bug#925473: tomcat9: sysvinit script missing (Policy §9.11¶2 “must”)

2019-04-01 Thread Thorsten Glaser
Gianfranco Costamagna 2013-2018, Emmanuel Bourg 2001-2017, Markus Koschany + 2015–2019, mirabilos License: Apache-2.0 License: Apache-2.0 diff --git a/debian/changelog b/debian/changelog index 9e1dab71..163eb8d9 100644 --- a/debian/changelog +++ b/debian/change

Bug#925929: LogsDirectory vs. group adm

2019-04-01 Thread Thorsten Glaser
On Mon, 1 Apr 2019, Felipe Sateler wrote: > Thanks for linking to the full file. I had not noticed that the unit used a > specific User. This means a root-owned /var/log/tomcat9 is not going to be > writable by tomcat. You should probably set it to tomcat9:adm, or add an Oops, yes, tomcat:adm is

Bug#926168: debian-policy: §9.3.2 difference to LSB (force-reload action of init scripts)

2019-04-01 Thread Thorsten Glaser
Package: debian-policy Version: 4.3.0.3 Severity: normal Hi, Policy 4.3.0.3 §9.3.2 reads: force-reload cause the configuration to be reloaded if the service supports this,

Bug#925929: LogsDirectory vs. group adm

2019-04-01 Thread Thorsten Glaser
Hi again Felipe, > If you ship this, there is no need for a LogsDirectory= entry. But I probably do need to add it with ReadWritePaths if we use ProtectSystem=strict, correct? https://salsa.debian.org/java-team/tomcat9/commit/5556481b345049f32720e20d22a072ebd9b865fa Thanks, //mirabilos --

Bug#925929: LogsDirectory vs. group adm

2019-03-29 Thread Thorsten Glaser
On Fri, 29 Mar 2019, Felipe Sateler wrote: > It might be a good idea to store that script somewhere else (possibly > /usr/lib/tomcat9 ?) and call it from the init script. As the script gets Indeed… the systemd unit calls the scripts from /usr/libexec/tomcat9/ (which is the currently recommended

Bug#925929: LogsDirectory vs. group adm

2019-03-29 Thread Thorsten Glaser
Hi Felipe, > > This won’t delete the logs on shutdown or something, because > > it’s called tmpfiles? > > No, because we don't provide the last argument (the age one): thanks! > In fact, /var/log is already tmpfile'd with a similar entry: OK. I don’t run systemd, so I didn’t know and needed

Bug#925929: LogsDirectory vs. group adm

2019-03-29 Thread Thorsten Glaser
Hi Felipe, > You can ship a tmpfiles snippet like: > > d /var/log/tomcat9 2750 root adm - This won’t delete the logs on shutdown or something, because it’s called tmpfiles? > If you ship this, there is no need for a LogsDirectory= entry. Thanks, //mirabilos -- tarent solutions GmbH

Bug#925454: RM: tomcat8 -- ROM; Replaced by tomcat9

2019-03-29 Thread Thorsten Glaser
On Fri, 29 Mar 2019, Moritz Mühlenhoff wrote: > IOW, there's no hard requirement and the Debian Java team has compelling > reasons (very limited workforce etc.) I’ve been a member of the Debian Java team for work-related reasons for a while and have expressed willingness to work on the init

Bug#925929: LogsDirectory vs. group adm

2019-03-29 Thread Thorsten Glaser
Hi, how can we make it so that a service that uses LogsDirectory has its logs readable by group adm? There is “LogsDirectoryMode=750” which we could change to 2750, but no way to set the group to adm, and from what I’ve read, pre-creating the /var/log/tomcat9 (in this case) as 2750 tomcat:adm

Bug#925929: tomcat9: /var/log/tomcat9 is owned by tomcat:tomcat instead of tomcat:adm and not setgid

2019-03-29 Thread Thorsten Glaser
On Fri, 29 Mar 2019, Emmanuel Bourg wrote: > The setgid bit is a good idea, otherwise the log files are only readable > by the tomcat group, and the adm group can't access them. In Debian at Indeed. Does this mean you agree with the patch I pushed to the git branch? > least exim4, mariadb, salt

Bug#925928: tomcat9: Error occurred during initialization of VM: Option -XX:+UseG1GC not supported

2019-03-29 Thread Thorsten Glaser
Emmanuel Bourg dixit: >Indeed, but running Tomcat on Zero is not a good idea anyway. ;-) In the end, we give users rope. Whether they hang themselves with it or use it to climb a rock is up to them. (Having the *ability* to do it, even if it's not *generally* a good idea, may be worth

Bug#925929: tomcat9: /var/log/tomcat9 is owned by tomcat:tomcat instead of tomcat:adm

2019-03-28 Thread Thorsten Glaser
Emmanuel Bourg dixit: >It's different from tomcat8, why do you think it is wrong? Because it’s customary, on Debian, that a user who is member of the group adm can read logs. (I’ve not checked whether this is an actual policy or where.) bye, //mirabilos -- “ah that reminds me, thanks for the

Bug#925928: tomcat9: Error occurred during initialization of VM: Option -XX:+UseG1GC not supported

2019-03-28 Thread Thorsten Glaser
Emmanuel Bourg dixit: >I can't reproduce this issue. What JRE did you use? G1 has been >available since Java 7. 11. Apparently, the option is not available in the Zero VM, only in Hotspot (didn’t check DCE, JamVM, etc). But the manpage says that a suitable GC is nowadays picked automatically

Bug#925929: tomcat9: /var/log/tomcat9 is owned by tomcat:tomcat instead of tomcat:adm and not setgid

2019-03-28 Thread Thorsten Glaser
retitle 925929 tomcat9: /var/log/tomcat9 is owned by tomcat:tomcat instead of tomcat:adm and not setgid thanks On Thu, 28 Mar 2019, Thorsten Glaser wrote: > commit aeff1188eaab8dbe99bb1fec62472d9c2ff1d876 changed the way > the log directory was created, and (probably a pasto) uses the &

Bug#925929: tomcat9: /var/log/tomcat9 is owned by tomcat:tomcat instead of tomcat:adm

2019-03-28 Thread Thorsten Glaser
Package: tomcat9 Version: 9.0.16-3 Severity: important commit aeff1188eaab8dbe99bb1fec62472d9c2ff1d876 changed the way the log directory was created, and (probably a pasto) uses the wrong ownership. -- System Information: Debian Release: buster/sid APT prefers unreleased APT policy: (500,

Bug#925928: tomcat9: Error occurred during initialization of VM: Option -XX:+UseG1GC not supported

2019-03-28 Thread Thorsten Glaser
Package: tomcat9 Version: 9.0.16-3 Severity: important tomcat9 does not start at all: […] NOTE: Picked up JDK_JAVA_OPTIONS: --add-opens=java.base/java.lang=ALL-UNNAMED --add-opens=java.base/java.io=ALL-UNNAMED --add-opens=java.rmi/sun.rmi.transport=ALL-UNNAMED Error occurred during

Bug#925621: virsh: error: internal error: Failed to probe QEMU binary with QMP:\n

2019-03-27 Thread Thorsten Glaser
On Wed, 27 Mar 2019, Thorsten Glaser wrote: > I cannot start any VM any more: > > tglase@tglase:~ $ virsh -c qemu:///system start Netboot > error: Failed to start domain Netboot > error: internal error: Failed to probe QEMU binary with QMP: I’ve just, after a reboot, seen someth

Bug#925621: virsh: error: internal error: Failed to probe QEMU binary with QMP:\n

2019-03-27 Thread Thorsten Glaser
Package: libvirt-clients Version: 5.0.0-1 Severity: important I cannot start any VM any more: tglase@tglase:~ $ virsh -c qemu:///system start Netboot error: Failed to start domain Netboot error: internal error: Failed to probe QEMU binary with QMP: 1|tglase@tglase:~ $ sudo tail -4

Bug#925440: Seriously?!

2019-03-26 Thread Thorsten Glaser
Gong S. dixit: >I'm on mobile, so forgive my formatting. No. >You should not require yet another hidden file in the user's home directory. You can set the environment variable ENV to move the file, which is *also* documented (and, incidentally, the reason we cannot read an mkshrc file from two

Bug#925440: Are you sure?

2019-03-25 Thread Thorsten Glaser
Gong S. dixit: >No, you did not. I created "/etc/mkshrc" and it is ignored by "mksh". The file ~/.mkshrc, prepopulated from /etc/skel/.mkshrc, loads /etc/mkshrc if it is non-empty (in Debian and derivatives, as well as several other packagings). If your user account was created before the mksh

Bug#925473: Bug#925454: RM: tomcat8 -- ROM; Replaced by tomcat9

2019-03-25 Thread Thorsten Glaser
On Mon, 25 Mar 2019, Emmanuel Bourg wrote: > When it's ready please let me review the update before uploading. OK. > Not changing the maintainer scripts nor the configuration files in the > process would be nice. I agree; at the current point in time, this should be least-intrusive. bye,

Bug#925454: RM: tomcat8 -- ROM; Replaced by tomcat9

2019-03-25 Thread Thorsten Glaser
On Mon, 25 Mar 2019, Emmanuel Bourg wrote: > We've got two serious security issues in the Tomcat init script in the You told me this. I replied by: • I’m active and willing to maintain it • if I don’t realise there’s a bug you can ping me • if even that fails, you can *still* remove it You

Bug#925454: RM: tomcat8 -- ROM; Replaced by tomcat9

2019-03-25 Thread Thorsten Glaser
clone 925454 -1 reassign -1 tomcat9 found -1 9.0.16-3 retitle -1 tomcat9: sysvinit script missing (Policy §9.11¶2 “must”) severity -1 serious tags -1 = confirmed outlook -1 working on it within the week --mirabilos owner -1 t...@mirbsd.de thanks On Mon, 25 Mar 2019, Scott Kitterman wrote: >

Bug#925454: RM: tomcat8 -- ROM; Replaced by tomcat9

2019-03-25 Thread Thorsten Glaser
On Mon, 25 Mar 2019, Emmanuel Bourg wrote: > Hi Thorsten, > > Le 25/03/2019 à 14:17, Thorsten Glaser a écrit : > > > Please don’t, tomcat9 has no sysvinit script, and you did > > not like any of my suggestions for it. > > I don't mind if tomcat8 is released wi

Bug#925454: RM: tomcat8 -- ROM; Replaced by tomcat9

2019-03-25 Thread Thorsten Glaser
On Mon, 25 Mar 2019, Emmanuel Bourg wrote: > Please remove the tomcat8 package from testing and unstable (not > experimental). This package has been replaced by tomcat9 for Buster. Please don’t, tomcat9 has no sysvinit script, and you did not like any of my suggestions for it. bye, //mirabilos

Bug#925407: openjdk-8: patches to make it build on older releases

2019-03-24 Thread Thorsten Glaser
Source: openjdk-8 Version: 8u212-b01-1 Severity: wishlist Tags: patch Hi doko, I’ve rebuilt the latest openjdk-8 on precise/trusty/xenial and wheezy/jessie (seeing you’re providing the newer releases with updates already) and found two issues (besides the testsuite taking ages or JAVA_HOME not

Bug#925359: dietlibc: built program on x32 terminates with 'smashed stack detected, program terminated.'

2019-03-24 Thread Thorsten Glaser
Bernhard Übelacker dixit: >I see that the syscall number gets modified to become 0x4062. > >But the syscall modifies 144 bytes, more than just the size of >variable ru1 of 88 bytes. > >This 144 bytes is the size I could observe within amd64 userland. The x32 syscalls often have struct

Bug#925359: dietlibc: built program on x32 terminates with 'smashed stack detected, program terminated.'

2019-03-23 Thread Thorsten Glaser
Bernhard Übelacker dixit: >Now I wonder if a x32 binary doing a syscall getrusage to a 64bit >kernel is supposed to supply memory like "long" would be 8 bytes? The x32 kernel is amd64, it’s just a userspace ABI, with a separate entry point in the kernel (OR’d a high bit in the syscall number).

Bug#925379: linux-base: linux-version vmlinuz/vmlinux dichotomy, breaks at least m68k

2019-03-23 Thread Thorsten Glaser
Package: linux-base Version: 4.5 Severity: important Control: affects -1 initramfs-tools Some time between 4.1.0-2 and 4.19.0-3, m68k kernels switched from vmlinuz-* to vmlinux-*: -rwxr-xr-x 1 root root 5933276 Feb 11 16:55 vmlinux-4.19.0-3-m68k* -rwxr-xr-x 1 root root 5933284 Mar 15 02:16

Bug#925358: klibc: built program on m68k terminates with signal 9 (SIGKILL)

2019-03-23 Thread Thorsten Glaser
Hi Adrian, >Could you point me to the actual error message so I know how the reproducer >looks like? sure: it’s the Total failed: 3 (3 unexpected) in the build log, and before that: FAIL ../../debian/mtest.t:mtest-external Description: Minitest: can run an

Bug#925359: dietlibc: built program on x32 terminates with 'smashed stack detected, program terminated.'

2019-03-23 Thread Thorsten Glaser
Package: dietlibc-dev Version: 0.34~cvs20160606-10 Severity: important https://buildd.debian.org/status/fetch.php?pkg=mksh=x32=57-1=1551461619=0 mksh built, with -fstack-protector-strong though, against dietlibc is not usable at all, on x32 only. -- System Information: Debian Release:

Bug#925358: klibc: built program on m68k terminates with signal 9 (SIGKILL)

2019-03-23 Thread Thorsten Glaser
Package: libklibc-dev Version: 2.0.6-1 Severity: important https://buildd.debian.org/status/fetch.php?pkg=mksh=m68k=57-1=1551464062=0 mksh built against klibc is not usable at all (thankfully, the build process recognises this and removes the unusable program, unless it is cross-built). Note

Bug#925299: /usr/bin/rmadison: rmadison: lies about LTS and other security uploads

2019-03-22 Thread Thorsten Glaser
Package: devscripts Version: 2.19.4 Severity: normal File: /usr/bin/rmadison Please reassign to the data source (and mark as affects devscripts, so the bug can still be found): (sid)$ rmadison libxml2 libxml2| 2.9.1+dfsg1-5 | oldstable-kfreebsd | source, kfreebsd-amd64,

Bug#816717: apt: fails to “download” local files saying it lacks permissions despite having them

2019-03-21 Thread Thorsten Glaser
retitle 816717 apt: fails to “download” local files saying it lacks permissions despite having them severity 910079 important merge 816717 910079 found 816717 1.8.0 thanks More on this: E: Failed to fetch file:/home/tglase/Projekte/DebTarent/dists/sid/tarent/binary-x32/Packages.gz Could not

Bug#925115: docker: Error response from daemon: unable to find "systemd" in controller set: unknown.

2019-03-19 Thread Thorsten Glaser
Package: docker.io Version: 18.09.1+dfsg1-5+b10 Severity: important Docker does not work out of the box. $ sudo docker run -it debian /bin/bash docker: Error response from daemon: unable to find "systemd" in controller set: unknown. Workaround: do this

Bug#686895: initscripts: /forcefsck: fsck -f undefined (e2fsck-ism)

2019-03-13 Thread Thorsten Glaser
On Wed, 13 Mar 2019, Dmitry Bogatov wrote: > Sorry, I am totally confused. Mind to make a patch? Eh, just don’t drop support, and at best document that people should use the new tune2fs option for ext[234]. No sense in making a patch, we’ve frozen. bye, //mirabilos -- «MyISAM tables -will-

Bug#906144: libmypaint-common: drop Conflicts: mypaint-data

2019-03-13 Thread Thorsten Glaser
severity 906144 serious reassign 906144 libmypaint-common found 906144 1.3.0-2.1 thanks As documented in #894757, gimp and mypaint not being coïnstallable is clearly RC. I’ve just poked around: it’s *only* the libmypaint.mo files that are affected, that is, the translation files of the

Bug#924441: inetutils-syslogd: world-readable logfiles after log rotation

2019-03-12 Thread Thorsten Glaser
Package: inetutils-syslogd Version: 2:1.9.4-7 I was a bit surprised to see that the logfiles created by logrotate on behalf of the inetutils-syslogd configuration are world-readable — on Debian, 0640 root:adm was expected. This might be deliberate, but I’ve not seen it documented explicitly.

Bug#686895: initscripts: /forcefsck: fsck -f undefined (e2fsck-ism)

2019-03-11 Thread Thorsten Glaser
On Mon, 11 Mar 2019, Dmitry Bogatov wrote: > So, you propose that we: > * drop all checks for /forcecheck No! > * document this fact in NEWS file > * write documentation, that e[2-4]fs users should use tune2fs tool >instead Yes. //mirabilos -- tarent solutions GmbH Rochusstraße 2-4,

Bug#686895: initscripts: /forcefsck: fsck -f undefined (e2fsck-ism)

2019-03-09 Thread Thorsten Glaser
On Sat, 9 Mar 2019, Pierre Ynard wrote: > /fastboot and /forcefsck are created by `shutdown -f` and `shutdown -F` Oh, I didn’t know that, I just sudo touch them as used to on Unix. > the shutdown binary, it could create /forcefsck or /run/forcefsck, which /run is a tmpfs. > would then be

Bug#686895: initscripts: /forcefsck: fsck -f undefined (e2fsck-ism)

2019-03-09 Thread Thorsten Glaser
On Sat, 9 Mar 2019, Dmitry Bogatov wrote: > As you convincingly remarked below, we may want honor `test -f /forcecheck' > for every filesystem, whose `fsck' supports `-f' option. As far as I > follow the thread, it is: > > ext2 ext3 ext4 reiserfs Interesting that acceptance of that

Bug#913119: Bug#913138: Bug#913119: linux-image-4.18.0-2-amd64: Hangs on lvm raid1

2019-03-06 Thread Thorsten Glaser
Hi Cesare, > But, in the meantime, doesn't the workaround of disabling blk_mq work > for you? I don’t know, because I have no means to trigger the bug. The three other virtualisators I’ve upgraded at the same time did not trigger it. My desktop machine also hits it only occasionally. I’ve

Bug#923795: /etc/cron.daily/tomcat7: compresses “live” logfiles

2019-03-05 Thread Thorsten Glaser
Package: tomcat7 Version: 7.0.56-3+really7.0.91-1 Severity: important /etc/cron.daily/tomcat7 compresses logfiles that are still “live” and just happen to not have been written to for a day. The next write to the still-opened logfile will go to nirvana, and furthermore, the second run fails

Bug#923559: systemd-udevd: writes to kernel log

2019-03-03 Thread Thorsten Glaser
log over kmsg for auto logging. (Closes: #923559) + + -- Thorsten Glaser Sun, 03 Mar 2019 23:00:57 +0100 + systemd (241-1) unstable; urgency=medium [ Adam Borowski ] diff -Nru systemd-241/debian/patches/debian/Use-syslog-for-AUTO-logging-before-klog.patch systemd-241/debian/patches/debi

Bug#923678: libjs-leaflet-markercluster: not a transitional package for apt

2019-03-03 Thread Thorsten Glaser
Package: libjs-leaflet-markercluster Version: 1.4.1~dfsg-5 Severity: important When removing the libjs-leaflet-markercluster “transitional” package, apt also clears the new one for removal: $ sudo apt-get purge libjs-leaflet-markercluster Reading package lists... Done Building dependency tree

Bug#923559: closed by Michael Biebl (Re: Bug#923559: systemd-udevd: writes to kernel log)

2019-03-01 Thread Thorsten Glaser
reopen 923559 thanks Debian Bug Tracking System dixit: >If this explanation is unsatisfactory and you have not received a >better one in a separate message then please contact Michael Biebl > by >replying to this email. This is unsatisfactory. >>This is expected behaviour. udevd uses the

Bug#923559: systemd-udevd: writes to kernel log

2019-03-01 Thread Thorsten Glaser
Package: udev Version: 241-1 Severity: normal Mar 1 23:09:44 tglase-nb vmunix: [ 125.926361] systemd-udevd[2636]: link_config: autonegotiation is unset or enabled, the speed and duplex are not writable. I expect this instead: Mar 1 23:09:44 tglase-nb systemd-udevd[2636]: link_config:

Bug#918368: do not parse apt output (was Re: [Git][pbuilder-team/pbuilder][master] 2 commits: buildpackage-funcs: change source of the default gcc.)

2019-02-28 Thread Thorsten Glaser
Mattia Rizzolo dixit: >buildpackage-funcs: change source of the default gcc. > >Thanks to Helmut for the suggestion. This breaks: # apt-cache show gcc | sed 's/^Depends: .*gcc-\([0-9]\+\) .*/\1/;t;d' 8 6 Don’t do that. Or, at least, ensure you only catch the installed version; the tool for

Bug#923481: alpine: replies lose In-Reply-To and References headers

2019-02-28 Thread Thorsten Glaser
Package: alpine Version: 2.21+dfsg1-1.1 Severity: serious If I take a message, reply to it, then go to the Subject line, press ^K to remove the existing (possibly damaged) text and type new text (possibly to change the subthread subject), the eMail gets sent out AND Fcc’d without both In-Reply-To

Bug#923478: initscripts use unsafe `: >` shell command to create files

2019-02-28 Thread Thorsten Glaser
Hi Pierre, I’d say that aborting might be preferrable, but I checked your patch. In the first case, it’s indeed preferrable to continue the init script (and have the two additional error messages from chmod and chgrp, but it continue on), and in the second case… >Init scripts try to use this

Bug#913119: Bug#913138: Bug#913119: linux-image-4.18.0-2-amd64: Hangs on lvm raid1

2019-02-28 Thread Thorsten Glaser
Hi Debian Linux kernel maintainers, On Wed, 27 Feb 2019, Dragan Milenkovic wrote: > > Hello Dragan, do you know if the patch was eventually included upstream and > > possibly in which version? > > Hello, Cesare. It was included in 4.20.11 and 4.19.24. please do consider uploading 4.19.24 to

Bug#904018: gcc-8: FTBFS on x32: hangs in the testsuite

2019-02-27 Thread Thorsten Glaser
found 904018 8.3.0-2 thanks https://buildd.debian.org/status/fetch.php?pkg=gcc-8=x32=8.3.0-2=1551277939=0 I’ve again uploaded a “nocheck” build, to let the game continue, but this ought to be eventually fixed. Thanks, //mirabilos -- «MyISAM tables -will- get corrupted eventually. This is a

Bug#923258: pbuilder-satisfydepends-apt: fails to consider libpam-systemd installed sometimes

2019-02-26 Thread Thorsten Glaser
On Mon, 25 Feb 2019, Thorsten Glaser wrote: > I’m trying to build texlive-extra (= 2018.20190131-2) (but this also > happens with -1), and after dependencies are installed, apt goes like: Same with src:tomcat8.

Bug#913119: linux-image-4.18.0-2-amd64: Hangs on lvm raid1

2019-02-26 Thread Thorsten Glaser
severity 913119 serious found 913119 4.19.16-1 thanks On Wed, 13 Feb 2019, Dragan Milenkovic wrote: > Jens Axboe has determined that the proper fix is quite different: > > http://git.kernel.dk/cgit/linux-block/commit/?h=for-linus=85bd6e61f34dffa8ec2dc75ff3c02ee7b2f1cbce > > This patch is

Bug#923272: linux: 50 s delay during boot [m68k, atari, ARAnyM]

2019-02-25 Thread Thorsten Glaser
Package: src:linux Version: 4.19.20-1 Severity: normal linux-image-4.19.0-3-m68k takes massively longer to boot than linux-image-4.1.0-2-m68k did (on an ARAnyM VM Atari using NatFeat network/disc/…): -BEGIN boot log of linux-image-4.19.0-3-m68k- Running Aranym headlessly ARAnyM 1.0.2

Bug#923258: pbuilder-satisfydepends-apt: fails to consider libpam-systemd installed sometimes

2019-02-25 Thread Thorsten Glaser
On Mon, 25 Feb 2019, Thorsten Glaser wrote: > I’m trying to build texlive-extra (= 2018.20190131-2) (but this also > happens with -1), and after dependencies are installed, apt goes like: […] > Temporarily changing /etc/pbuilderrc to set > PBUILDERSATISFYDEPENDSCMD=/usr/lib/pbuil

Bug#923258: pbuilder-satisfydepends-apt: fails to consider libpam-systemd installed sometimes

2019-02-25 Thread Thorsten Glaser
Package: pbuilder Version: 0.230.1 Severity: normal This is a really weird bug. I’m in a sid/amd64 chroot with sysvinit installed (basically, upgraded from, I think, a lenny/amd64 chroot to sid then kept updated all the time). There are no held packages inside the chroot, no APT pinning, etc. (I

Bug#920459: tabu: breaks with color and/or xcolor when spread or negative X coefficients are used

2019-02-24 Thread Thorsten Glaser
Hi Norbert, >Go ahead immediately, you have every approval. >BTW, changes can be sent to the new development place > https://github.com/debian-tex/texlive-nonbin OK, will do so. I’ll give people a bit to review, although someone on TeX.SE said tabu is broken enough there’s no way I can

Bug#920459: tabu: breaks with color and/or xcolor when spread or negative X coefficients are used

2019-02-24 Thread Thorsten Glaser
) + + -- Thorsten Glaser Sun, 24 Feb 2019 17:47:26 +0100 + texlive-extra (2018.20190131-1) unstable; urgency=medium * new upstream checkout diff -Nru texlive-extra-2018.20190131/debian/patches/fix-tabu-920459 texlive-extra-2018.20190131/debian/patches/fix-tabu-920459 --- texlive-extra

Bug#914632: uw-imap: CVE-2018-19518

2019-02-24 Thread Thorsten Glaser
Hi Magnus, >Perhaps wanting to run imapd via remote shell is so rare that there's >no need to write a NEWS.Debian entry? in case of doubt just write one, it does not hurt. Are you going to upload within the next five days or so, or do you need help? (We’re at a BSP and currently fixing stuff…)

Bug#923067: libc0.1: missing UTIME_OMIT for utimensat(2) and futimens(2)

2019-02-24 Thread Thorsten Glaser
On Sat, 23 Feb 2019, Thorsten Glaser wrote: > then give-back src:pax on both kfreebsd architectures. In the meantime, I had to do a pax upload to fix bugs anyway, so I just added a configure-time check for UTIME_OMIT, so it built now, although not using utimensat/futimes. In case you won

Bug#923098: libklibc-dev: missing clock_gettime

2019-02-23 Thread Thorsten Glaser
Package: libklibc-dev Version: 2.0.6-1 Severity: wishlist Tags: upstream Hi, clock_gettime(2) support (with at least CLOCK_REALTIME (although I guess gettimeofday(2) is the same just with less precision) and CLOCK_MONOTONIC) would be nice for something I’m working on. -- System Information:

Bug#923073: screen: chdir does not do tilde substitution

2019-02-23 Thread Thorsten Glaser
Package: screen Version: 4.6.2-3 Severity: minor ^A:chdir ~/foo/bar does not work. ^A:source ~/.screenrc-foo does work since recently, so it got fixed in at least some other place in the source, the fix now “only” needs to be applied everywhere else pathnames are used. -- System Information:

Bug#923067: libc0.1: missing UTIME_OMIT for utimensat(2) and futimens(2)

2019-02-23 Thread Thorsten Glaser
Package: libc0.1 Version: 2.25-2 Severity: normal glibc on Debian GNU/kFreeBSD is missing UTIME_OMIT, making src:pax FTBFS. While I could work around this by using the older functions, https://www.freebsd.org/cgi/man.cgi?query=utimensat indicates that the FreeBSD kernel has gained native support

Bug#914632: uw-imap: CVE-2018-19518

2019-02-23 Thread Thorsten Glaser
Hi Magnus, >I reckon. I just haven't been able to make gbp use my long PGP key id... any progress with that? Otherwise I’d be willing to NMU your patch. Greetings from the BSP, //mirabilos -- 21:12⎜ sogar bei opensolaris haben die von der community so ziemlich jeden mist eingebaut │ man sollte

Bug#922990: dh_installinitramfs: causes script-not-executable in lintian

2019-02-22 Thread Thorsten Glaser
Package: debhelper Version: 12.1 Severity: important dh_installinitramfs does not set the scripts it installs into debian/PACKAGE/usr/share/initramfs-tools/hooks/ executable. Worse, it removes the executable bit even if it was set in debian/PACKAGE.initramfs-hook before. (But it should not need

Bug#918590: WARNING: Device /dev/md0 not initialized in udev database even after waiting 10000000 microseconds

2019-02-21 Thread Thorsten Glaser
On Thu, 21 Feb 2019, Axel Beckert wrote: > Another thing I need to check before being able to answer is if I > still have that "sleep 5" workaround in the udev init script which > Michael Biebl suggested back then and which at least helped with some > recent udev/LVM issues on some, but not all

  1   2   3   4   5   6   7   8   9   10   >