Bug#1021738: man2html: CVE-2021-40647 CVE-2021-40648

[Robert Luberda]

clone 1021738 -1
retitle 1021738 man2html: CVE-2021-40647
tags 1021738 +pending
retitle -1 man2html: CVE-2021-40648
tags -1 +moreinfo
thanks

Moritz Mühlenhoff pisze:

Hi

First of all I'm sorry for not taking care about it earlier, I didn't 
have time for Debian work in the previous year.




The following vulnerabilities were published for man2html.

CVE-2021-40647[0]:

Ok, this is quite easy to fix, I will upload fixed version soon.


CVE-2021-40648[1]:
| In man2html 1.6g, a filename can be created to overwrite the previous
| size parameter of the next chunk and the fd, bk, fd_nextsize,


According to instructions given at 
https://gist.github.com/untaman/cb58123fe89fc65e3984165db5d40933 I tried 
to reproduce this with the following commands:

  file=$(perl -e 'print "A" x 132')
  touch $file
  man2html $file
I used man2html built with AddressSanitizer and it found only a few 
small memory leaks coming from global variables.


So I have no idea what really is wrong in this CVE. The source code 
references given at the above link actually refer to calls to 
fopen()/fclose() functions rather then to directly malloc() and free() 
directly.


Regards,
robert

Bug#1061270: lintian: Reports bogus bin-sbin-mismatch issues

[Robert Luberda]

Package: lintian
Version: 2.116.3
Severity: normal

For dwww package lintian reports:
X: dwww: bin-sbin-mismatch usr/sbin/dwww -> usr/bin/dwww [etc/cron.daily/dwww]
X: dwww: bin-sbin-mismatch usr/sbin/dwww -> usr/bin/dwww [etc/cron.weekly/dwww]
X: dwww: bin-sbin-mismatch usr/sbin/dwww -> usr/bin/dwww [postinst]
X: dwww: bin-sbin-mismatch usr/sbin/dwww -> usr/bin/dwww [usr/lib/cgi-bin/dwww]
X: dwww: bin-sbin-mismatch usr/sbin/dwww -> usr/bin/dwww 
[usr/sbin/dwww-refresh-cache]

And lintian is simply wrong here. The dwww package contains a  single 
/usr/bin/dwww command and a few other /usr/sbin/dwww-something commands.
All the scripts listed above check the commands correcly. For example 
/etc/cron-weekly/dwww
basically contains this (unimportant stuff removed):

  # check if swish++ is installed
  test -x /usr/bin/index++ || exit 0

  # check if dwww is still installed
  test -x /usr/sbin/dwww-index++ || exit 0

  # See ionice(1)
  [ -x /usr/bin/ionice ] &&  IONICE="/usr/bin/ionice -c3 -t" || IONICE=

  $IONICE dwww-index++ > /dev/null

Regards,
robert


-- System Information:
Debian Release: trixie/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'stable-security'), (990, 'unstable'), (990, 'stable'), (500, 
'oldstable-updates'), (500, 'oldstable-security'), (500, 'oldstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 6.6.11-amd64 (SMP w/4 CPU threads; PREEMPT)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages lintian depends on:
ii  binutils2.41.90.20240115-1
ii  bzip2   1.0.8-5+b2
ii  diffstat1.65-1
ii  dpkg1.22.2
ii  dpkg-dev1.22.2
ii  file1:5.45-2+b1
ii  gettext 0.21-14
ii  gpg 2.2.40-1.1+b1
ii  intltool-debian 0.35.0+20060710.6
ii  iso-codes   4.16.0-1
ii  libapt-pkg-perl 0.1.40+b3
ii  libarchive-zip-perl 1.68-1
ii  libberkeleydb-perl  0.64-2+b2
ii  libcapture-tiny-perl0.48-2
ii  libclass-xsaccessor-perl1.19-4+b2
ii  libclone-perl   0.46-1+b1
ii  libconfig-tiny-perl 2.30-1
ii  libconst-fast-perl  0.014-2
ii  libcpanel-json-xs-perl  4.37-1+b1
ii  libdata-dpath-perl  0.59-1
ii  libdata-validate-domain-perl0.10-1.1
ii  libdata-validate-uri-perl   0.07-2
ii  libdevel-size-perl  0.83-2+b2
pn  libdigest-sha-perl  
ii  libdpkg-perl1.22.2
ii  libemail-address-xs-perl1.05-1+b2
ii  libfile-basedir-perl0.09-2
ii  libfile-find-rule-perl  0.34-3
ii  libfont-ttf-perl1.06-2
ii  libhtml-html5-entities-perl 0.004-3
ii  libhtml-tokeparser-simple-perl  3.16-4
ii  libio-interactive-perl  1.025-1
ii  libipc-run3-perl0.048-3
ii  libjson-maybexs-perl1.004005-1
ii  liblist-compare-perl0.55-2
ii  liblist-someutils-perl  0.59-1
ii  liblist-utilsby-perl0.12-2
ii  libmldbm-perl   2.05-4
ii  libmoo-perl 2.005005-1
ii  libmoox-aliases-perl0.001006-2
ii  libnamespace-clean-perl 0.27-2
ii  libpath-tiny-perl   0.144-1
ii  libperlio-gzip-perl 0.20-1+b2
ii  libperlio-utf8-strict-perl  0.010-1+b1
ii  libproc-processtable-perl   0.636-1+b1
ii  libregexp-wildcards-perl1.05-3
ii  libsereal-decoder-perl  5.004+ds-1+b1
ii  libsereal-encoder-perl  5.004+ds-1+b1
ii  libsort-versions-perl   1.62-3
ii  libsyntax-keyword-try-perl  0.29-1+b1
ii  libterm-readkey-perl2.38-2+b2
ii  libtext-levenshteinxs-perl  0.03-5+b2
ii  libtext-markdown-discount-perl  0.16-1+b1
ii  libtext-xslate-perl 3.5.9-1+b3
ii  libtime-duration-perl   1.21-2
ii  libtime-moment-perl 0.44-2+b2
ii  libtimedate-perl2.3300-2
ii  libunicode-utf8-perl0.62-2+b1
ii  liburi-perl 5.21-1
ii  libwww-mechanize-perl   2.17-1
ii  libwww-perl 6.73-1
ii  libxml-libxml-perl  2.0207+dfsg+really+2.0134-1+b2
ii  libyaml-libyaml-perl0.86+ds-1+b1
ii  lzip [lzip-decompressor]1.23-6
ii  lzop1.04-2
ii  man-db  2.12.0-3
ii  patchutils  0.4.2-1
ii  perl [libencode-perl]   5.38.2-3
ii  t1utils 1.41-4
ii  unzip   6.0-28
ii  xz-utils5.4.5-0.3

lintian recommends no packages.

Versions of packages lintian suggests:
pn  binutils-multiarch 
ii  

Bug#1061271: reportbug: Please make "Selected mail user agent cannot be found" more user friendly

[Robert Luberda]

Package: reportbug
Version: 12.0.0
Severity: wishlist

Today I've tried to run reportbug after a pretty long time of not using
it, and it turned out that it did nothing, but printing something like:
"Selected mail user agent cannot be found, exiting".

It took me a few minutes to discover that I have ~/.reportbugrc from
2012 containing a "mutt" line, but mutt package was no longer installed
on my system (most likely it got removed some time ago by mistake).
After reinstalling mutt, reportbug started working fine.

But I wish the message printed by reportbug in such cases would be more
helpful. It would be nice if it could at least include the name of the
selected mua, for example: "Selected mail user agent 'mutt' cannot be
found". (It would be even greater if it could print the name of the
config file as well, e.g. "Mail user agent 'mutt' selected in
~/.reportbugrc cannot be found").

Regards,
robert

-- Package-specific info:
** Environment settings:
EDITOR="vim"
DEBEMAIL="rob...@debian.org"
DEBFULLNAME="Robert Luberda"
INTERFACE="text"

** /home/robert/.reportbugrc:
reportbug_version "1.99.29"
mode advanced
ui text
realname "Robert Luberda"
email "rob...@debian.org"
no-cc
offline
mutt

-- System Information:
Debian Release: trixie/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'stable-security'), (990, 'unstable'), (990, 'stable'), (500, 
'oldstable-updates'), (500, 'oldstable-security'), (500, 'oldstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 6.6.11-amd64 (SMP w/4 CPU threads; PREEMPT)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages reportbug depends on:
ii  apt2.7.10
ii  python33.11.6-1
ii  python3-reportbug  12.0.0
ii  sensible-utils 0.0.20

reportbug recommends no packages.

Versions of packages reportbug suggests:
pn  claws-mail  
ii  debconf 1.5.83
ii  debsums 3.0.2.1
ii  dlocate 1.12
pn  emacs-bin-common
ii  file1:5.45-2+b1
ii  gnupg   2.2.40-1.1
ii  postfix [mail-transport-agent]  3.8.4-1
pn  python3-urwid   
pn  reportbug-gtk   
ii  xdg-utils   1.1.3-4.1

Versions of packages python3-reportbug depends on:
ii  apt2.7.10
ii  file   1:5.45-2+b1
ii  python33.11.6-1
ii  python3-apt2.7.5
ii  python3-debian 0.1.49
ii  python3-debianbts  4.0.2
ii  python3-requests   2.31.0+dfsg-1
ii  sensible-utils 0.0.20

python3-reportbug suggests no packages.

-- no debconf information

Bug#1025452: lintian: useless checks of debian/tests datafiles

[Robert Luberda]

Package: lintian
Version: 2.115.3
Severity: important

I'm working on new upstream version of upx-ucl. The package contains
some *intentionally broken* ELF files inside debian/tests to check
that for example upx does not crash on them. However for some reason 
lintan wants the data files to be perfectly correct binaries, with an 
accompanying source files. IMHO this is just ridiculous. Those files
are in debian/tests anyway, what's the point of having such strict
requirements on them?

Anyway, some time ago I added overrides for those warnings, 
but now I can see that:
- the overrides do no longer work, because lintian now requires
  names to be put inside square brackets;
- some new checks were implemented to see if ELF file is correct,
  and the new checks display a bunch of additional warnings making it
  completely unreadable (and I do not think they can be overridden).

Please just look at this lintian output of upx-ucl 3.96-3 package:

Now running lintian --display-info --display-experimental --pedantic 
--color=always upx-ucl_3.96-3_amd64.changes ...
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 1]:   [ 0]   00010102:  0001003e0002 405418 
40 43  0   0 15762873573703680 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 9]:   [ 4]   00ce:  00010102464c457f 
4025b00d003e0002 35166de4b60f d84f21364027050f A 
51577028 955743702 432376351040014592 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 11]:   [ 5]   27080001:  213c84c840070f02 
ec0ec27011c 2fb3cc074005 7b0b90641704e460 WAxMILTCo 9475801 
2872516608 510523112357773942 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 15]:   [ 8]   0400322e:  0302dd34cfba 
fff60303141f2006 5631d867221fc1ff 490437432c17f6f7 ox 2146830080 
2207070372 1972876281791746561 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 19]:   [10]   2f311b08:  086c21b017098f12 
864219082f6247e5 17866c21b0865e62 a7021b086c212f29 AMLOGTop 1691277137 
398596360 12089950241734945530 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 22]:   [11]   5f55b086:  f14d86421908bb17 
17488fe56c21b086 745f20421b086198 2b02d0759b23b0 WAXMSLOGTCxxlExx 
1814138935 2486117640 3417519580065700616 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 24]:   [12]   17dd86c2:  01804219086c0b2f 
8a1d42190864217 86373677841b0861 432f5b21b0864277 WAMSLGTxop 527180208 
3944157743 6210289578152757356 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 26]:   [13]   1bc3178d:  701738086c21902f 
2f8c5f7b86c21b08 1748bf3d6c21b086 21039f0002a87272 AILGTxop 3076108358 1132526 
4860816968149843738 at /usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 
569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 29]:   [14]   2fb31817:  17206c87272077ac 
4d862689f100039 5fff0036de63efa8 746573006b68635f WAXMop 1718026210 
1752397164 7382801058585599744 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 34]:   [18]   5f6f6eb5:  adbe103fc928f28c 
64742666807a4b04 1b74845f6ec7756f 6e3eddf0a16d956d XxSIOTCop 1699376211 
2037396843 8452239821877519578 at 
/usr/share/lintian/bin/../lib/Lintian/Index/Elf.pm line 569.
upx-ucl_3.96-3.dsc (patched): Warning while running readelf 
ondebian/tests/data/907426-poc_free: Parse error in readelf section headers 
[row 36]:   [19]   372f004f:  bdb7bdb77e62bd6f 
6567208368764a6d 7784009dba17c2b5 322738739d0cdd91 WxMILOTCxop 2826200169 
3671557790 3259666543423467181 at 

Bug#1025131: apt-cacher-cleanup does not work

[Robert Luberda]

Mark Hindley pisze:

Hi,


Control: tags -1 patch

Robert,

Many thanks for pointing this out.

Does the attached patch help?


No, due to perl's syntax error related to unmatched parenthesis. I'm 
attaching a patch that actually works.


Regards,
Robert
From cc36b2f690d694ad0cfe1b1c0ca5df7ead3e57ca Mon Sep 17 00:00:00 2001
From: Mark Hindley 
Date: Wed, 30 Nov 2022 11:30:21 +
Subject: [PATCH] Add encoded underscores to *_files_regexp. Test fix for
 #1025131.

---
 lib/apt-cacher.pl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/lib/apt-cacher.pl b/lib/apt-cacher.pl
index 712d0bc..dad985a 100755
--- a/lib/apt-cacher.pl
+++ b/lib/apt-cacher.pl
@@ -132,7 +132,7 @@ sub read_config {
 			  qw(vmlinuz
 			 linux
 			 initrd\.gz
-			 (?:%VALID_PACKAGE_NAME%_%VALID_VERSION%[_\.])?changelog
+			 (?:%VALID_PACKAGE_NAME%(?:_|%5f)%VALID_VERSION%[_\.])?changelog
 			 NEWS\.Debian
 			 %VALID_UBUNTU_RELEASE_NAMES%\.tar\.gz(?:\.gpg)?
 			 (?:by-hash/(?i:MD5SUM/[0-9a-f]{32}|SHA1/[0-9a-f]{40}|SHA256/[0-9a-f]{64}))
@@ -141,7 +141,7 @@ sub read_config {
 			   )
 			   ) . ')$',
 		  package_files_regexp => '(?:' . join('|',
-		   qw((?:^|/)%VALID_PACKAGE_NAME%_%VALID_VERSION%(?:_%VALID_ARCHS%\.(?:u|d)?deb|\.dsc|\.tar\.(?:gz|bz2|xz|lzma)(?:\.asc)?|\.diff\.gz)
+		   qw((?:^|/)%VALID_PACKAGE_NAME%(?:_|%5f)%VALID_VERSION%(?:(?:_|%5f)%VALID_ARCHS%\.(?:u|d)?deb|\.dsc|\.tar\.(?:gz|bz2|xz|lzma)(?:\.asc)?|\.diff\.gz)
 			  \.rpm
 			  index\.db-.+\.gz
 			  \.jigdo
-- 
2.35.1

Bug#1025131: apt-cacher-cleanup does not work

[Robert Luberda]

Package: apt-cacher
Version: 1.7.27
Severity: important

The "Encode embedded underscores in URLs when building filenames"
change introduced in 1.7.27 broke apt-cacher-cleanup.pl: it does 
not clean *.deb files at all, as stored file names such as
 adduser%5f3.127%5fall.deb
 adduser%5f3.128%5fall.deb
 adduser%5f3.129%5fall.deb
do no longer match to the $package_files_regexp.

Regards
robert



-- System Information:
Debian Release: bookworm/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(500, 'stable-updates'), (500, 'stable-security'), (500, 'stable')
merged-usr: no
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 6.0.0-2-amd64 (SMP w/4 CPU threads; PREEMPT)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages apt-cacher depends on:
ii  debconf [debconf-2.0]  1.5.79
ii  distro-info-data   0.56
ii  ed 1.18-1
ii  init-system-helpers1.64
ii  libdpkg-perl   1.21.9
ii  libfilesys-df-perl 0.92-7+b1
ii  libio-interactive-perl 1.023-2
ii  libio-interface-perl   1.09-2+b2
ii  libipc-sharelite-perl  0.17-4+b7
ii  libnetaddr-ip-perl 4.079+dfsg-2+b1
ii  libsys-syscall-perl0.25-7
ii  libwww-curl-perl   4.17-8+b1
ii  libwww-perl6.67-1
ii  lsb-base   11.5
ii  perl   5.36.0-4
ii  sysvinit-utils [lsb-base]  3.05-6
ii  update-inetd   4.51

Versions of packages apt-cacher recommends:
ii  libberkeleydb-perl0.64-2+b1
ii  libio-compress-lzma-perl  2.201-1

Versions of packages apt-cacher suggests:
pn  libfreezethaw-perl   
ii  libio-socket-inet6-perl  2.73-1

-- Configuration Files:
/etc/apt-cacher/apache.conf changed [not included]
/etc/apt-cacher/apt-cacher.conf changed [not included]
/etc/cron.d/apt-cacher changed [not included]

-- debconf information excluded

-- debsums errors found:
debsums: changed file /usr/share/apt-cacher/apt-cacher-cleanup.pl (from 
apt-cacher package)
debsums: changed file /usr/share/apt-cacher/lib/apt-cacher.pl (from apt-cacher 
package)

Bug#1023501: busybox-static: version 1:1.35.0-3 breaks boot on hppa

[Robert Luberda]

severity 1023501 grave
retitle 1023501 busybox-static: version 1:1.35.0-3 breaks boot on hppa 
and amd64

found 1023501 1:1.35.0-3
notfound 1023501  1:1.35.0-2

On Sat, 05 Nov 2022 13:31:51 + John David Anglin 
 wrote:

Package: busybox-static
Version: 1:1.35.0-2
Severity: normal

Dear Maintainer,

With 1:1.35.0-3, boot ends in initramfs:

Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... done.
Begin: Waiting for root file system ... Begin: Running /scripts/local-block ... 
   
mdadm: No arrays found in config file or automatically
done.
Begin: Running /scripts/local-block ... mdadm: No arrays found in config file 
or
automatically


I had the same issue on amd64.
Removing mdadm package did not help.
Downgrading busybox-static to 1.35.0-2 fixed the issue.


I'm including the system information generated by reportbug below:


-- System Information:
Debian Release: bookworm/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 
'testing'), (500, 'stable-updates'), (500, 'stable-security'), (500, 
'stable')

merged-usr: no
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 6.0.0-2-amd64 (SMP w/4 CPU threads; PREEMPT)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), LANGUAGE 
not set

Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages busybox depends on:
ii  libc6  2.36-4

busybox recommends no packages.

busybox suggests no packages.

-- no debconf information

Regards,
Robert

Bug#1014730: liburi-perl: Breaks apt-cacher with "Can't locate Regexp/IPv6.pm" error

[Robert Luberda]

gregor herrmann pisze:


I'm a bit surprised, because the new code in 5.11 should behave
better if Regexp::IPv6 is not available:


Yes, but it looks like apt-cacher seems to set its own SIG{__DIE__} handler.



+our $IPv6_re;
+
+sub _looks_like_raw_ip6_address {
+  my $addr = shift;
+
+  if ( !$IPv6_re ) { #-- lazy / runs once / use Regexp::IPv6 if installed
+eval {


I've just checked that adding

  local $SIG{__DIE__};

here (i.e. before the require line) fixes the issue with apt-cacher for me.


+  require Regexp::IPv6;
+  Regexp::IPv6->import( qw($IPv6_re) );
+  1;
+}  ||  do { $IPv6_re = qr/[:0-9a-f]{3,}/; }; #-- fallback: unambitious 
guess




But may as well move libregexp-ipv6-perl to Depends, I guess.


Probably yes, but IMHO it would be better to restore default __DIE__ 
handler.


Regards,
robert

Bug#1014730: liburi-perl: Breaks apt-cacher with "Can't locate Regexp/IPv6.pm" error

[Robert Luberda]

Package: liburi-perl
Version: 5.11-1
Severity: serious
Justification: Policy 7.2
Control: affects -1 apt-cacher

After recent upgrade I've noticed that apt-cached does no longer work:
 Err:2 http://apt-testing.vox:3142/apt-cacher/debian bookworm InRelease 

   502  apt-cacher internal error (died) [IP: 127.0.0.1 3142]

The following information is printed into apt-cacher.log:
Mon Jul 11 00:02:52 2022|error [18513]: Can't locate Regexp/IPv6.pm in @INC 
(you may need to install the Regexp::
IPv6 module) (@INC contains: /usr/share/apt-cacher/lib /etc/perl 
/usr/local/lib/x86_64-linux-gnu/perl/5.34.0 /usr
 /local/share/perl/5.34.0 /usr/lib/x86_64-linux-gnu/perl5/5.34 /usr/share/perl5 
/usr/lib/x86_64-linux-gnu/perl-bas
e /usr/lib/x86_64-linux-gnu/perl/5.34 /usr/share/perl/5.34 
/usr/local/lib/site_perl) at /usr/share/perl5/URI/_gen
eric.pm line 25.

Downgrading liburi-perl to 5.10 fixes the issue.

Regards,
robert


-- System Information:
Debian Release: bookworm/sid
  APT prefers unstable-debug
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.18.0-2-amd64 (SMP w/4 CPU threads; PREEMPT)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages liburi-perl depends on:
ii  perl  5.34.0-5

liburi-perl recommends no packages.

Versions of packages liburi-perl suggests:
pn  libbusiness-isbn-perl  
pn  libregexp-ipv6-perl
ii  libwww-perl6.67-1

-- no debconf information

Bug#982566: procps: Fails to install

[Robert Luberda]

Package: procps
Version: 2:3.3.17-2
Severity: serious
Justification: file conflict

Hi,

procps fails to install due to file conflict with manpages-pl:

Unpacking procps (2:3.3.17-2) over (2:3.3.17-1) ...
dpkg: error processing archive 
/var/cache/apt/archives/procps_2%3a3.3.17-2_amd64.deb (--unpack):
 trying to overwrite '/usr/share/man/pl/man1/free.1.gz', which is also in 
package manpages-pl 1:4.2.0-1
dpkg: error while cleaning up:
 new procps package post-removal script subprocess returned error exit status 1
Errors were encountered while processing:
 /var/cache/apt/archives/procps_2%3a3.3.17-2_amd64.deb
mount: /usr: mount point is busy.
E: Sub-process /usr/bin/dpkg returned an error code (1)

Regards,
robert


-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'unstable'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Bug#981909: debci: ci.debian.net does not show all results

[Robert Luberda]

Package: debci
Severity: normal

Hi

My DDPO page contains a link to 
https://ci.debian.net/packages/s/sysstat/testing/amd64/ 
which currently has two build logs displayed in grey color (versions 12.5.2-2 
and 12.5.2-1
from unstable were tested against packages in testing):

  Version   Date   Trigger/Pinned packages  
  12.5.2-2  2021-02-03 09:37:11 UTCsysstat/12.5.2-2   
   src:sysstat from unstable
  12.5.2-1  2021-01-31 05:07:28 UTCsysstat/12.5.2-1
   src:sysstat from unstable

(note I stripped the other data to have the above lines shorter)

However the link to the above page is not shown on the the main page 
for sysstat at https://ci.debian.net/packages/s/sysstat/, which
currently contains:

  unstabletesting stable  oldstable
  amd64   12.5.2-1 fail   No test dataNo test dataNo test data
  arm64   12.5.2-1 fail   No test dataNo test dataNo test data
  armhf   No test dataNo test dataNo test dataNo test data
  i38612.5.2-1 fail   No test dataNo test dataNo test data
  ppc64el No test dataNo test dataNo test dataNo test data
  s390x   No test dataNo test dataNo test dataNo test data 

This means that to see for example results of run of systat 12.5.2 from
unstable against packages in testing for architectures other than amd64,
I need to manually change the link provided on my DDPO page. I can see
that such links point to valid pages, e.g. 
https://ci.debian.net/packages/s/sysstat/testing/i386/
https://ci.debian.net/packages/s/sysstat/testing/ppc64el/
so why not to provide access to them from the main index page for
package?

Regards,
robert

Bug#981890: O: apt-listchanges -- package change history notification tool

[Robert Luberda]

Package: wnpp
Severity: normal

I intend to orphan the apt-listchanges package. I don't have time to
properly maintain it. I will upload 3.23 shortly with maintainer set 
to Debian QA Group.

Regards,
robert



signature.asc
Description: PGP signature

Bug#981584: scons: MD5-timestamp decider does no longer follow symlinks

[Robert Luberda]

forwarded 981584 https://github.com/SCons/scons/issues/3880
thanks

Bill Deegan wrote:
> You don't need to be a scons-users subscriber to file the issue.

According to my past experience, the other scons teams members disagree
with you.

> I'm the project maintainer.
> You can just specify in the issue that I (bdbaddog) asked to you file it.

Ok, done.

Regards,
Robert

Bug#981584: scons: MD5-timestamp decider does no longer follow symlinks

[Robert Luberda]

Bill Deegan write:
> Also try against 4.1.0 (recently released)

I wasn't aware about new release, so thanks for the hint. But
unfortunately it does not fix the issue :(


> 
> On Mon, Feb 1, 2021 at 10:44 AM Bill Deegan  > wrote:
> 
> Any chance you can file this on SCons's github repo?

I would prefer not to, especially that I am not scons-users subscriber.

Regards,
robert

Bug#981584: scons: MD5-timestamp decider does no longer follow symlinks

[Robert Luberda]

Package: scons
Version: 4.0.1+dfsg-2
Severity: normal
Tags: upstream

Hi,

I'm attaching a simple test project that:
- contains hello2.c file that is symbolic link to hello.c file;
- and sets 'MD5-timestamp' as the decider function.

Previously (what I'm proving at the bottom of this mail) scons used
to build hello2 when hello.c was changed. In the current version of 
scons only hello gets rebuilt.

# Initial compilation with scons 4:
  robert@vox:/tmp/proj$ scons --debug=explain
  scons: Reading SConscript files ...
  scons: done reading SConscript files.
  scons: Building targets ...
  scons: building `hello.o' because it doesn't exist
  gcc -o hello.o -c hello.c
  scons: building `hello' because it doesn't exist
  gcc -o hello hello.o
  scons: building `hello2.o' because it doesn't exist
  gcc -o hello2.o -c hello2.c
  scons: building `hello2' because it doesn't exist
  gcc -o hello2 hello2.o
  scons: done building targets.

# The following command changes both hello.c and hello2.c...  
  robert@vox:/tmp/proj$ sed -i -e 's/word/WoRd/i' hello.c

# ... but hello2.c is not rebuilt:
  robert@vox:/tmp/proj$ scons --debug=explain
  scons: Reading SConscript files ...
  scons: done reading SConscript files.
  scons: Building targets ...
  scons: rebuilding `hello.o' because:
 `hello.c' changed
 `/usr/bin/gcc' changed
  gcc -o hello.o -c hello.c
  scons: rebuilding `hello' because:
 `hello.o' changed
 `/usr/bin/gcc' changed
  gcc -o hello hello.o
  scons: done building targets.

  robert@vox:/tmp/proj$ scons -v
  SCons by Steven Knight et al.:
SCons: v4.0.1.c289977f8b34786ab6c334311e232886da7e8df1, 2020-07-17 
01:50:03, by bdbaddog on ProDog2020
SCons path: ['/usr/lib/python3/dist-packages/SCons']
  Copyright (c) 2001 - 2020 The SCons Foundation



After downgrading scons to 3.1.2+dfsg-0.1, it works as I would expect:

# Initial compilation with scons 3:
  robert@vox:/tmp/proj$ scons --debug=explain
  scons: Reading SConscript files ...
  scons: done reading SConscript files.
  scons: Building targets ...
  scons: building `hello.o' because it doesn't exist
  gcc -o hello.o -c hello.c
  scons: building `hello' because it doesn't exist
  gcc -o hello hello.o
  scons: building `hello2.o' because it doesn't exist
  gcc -o hello2.o -c hello2.c
  scons: building `hello2' because it doesn't exist
  gcc -o hello2 hello2.o
  scons: done building targets.
  
# Still both files are changed by the following command:
  robert@vox:/tmp/proj$ sed -i -e 's/word/WORd/i' hello.c
# ... and both are rebuilt:
  robert@vox:/tmp/proj$ scons --debug=explain
  scons: Reading SConscript files ...
  scons: done reading SConscript files.
  scons: Building targets ...
  scons: rebuilding `hello.o' because:
 `hello.c' changed
 `/usr/bin/gcc' changed
  gcc -o hello.o -c hello.c
  scons: rebuilding `hello' because:
 `hello.o' changed
 `/usr/bin/gcc' changed
  gcc -o hello hello.o
  scons: rebuilding `hello2.o' because:
 `hello2.c' changed
 `/usr/bin/gcc' changed
  gcc -o hello2.o -c hello2.c
  scons: rebuilding `hello2' because:
 `hello2.o' changed
 `/usr/bin/gcc' changed
  gcc -o hello2 hello2.o
  scons: done building targets.

  robert@vox:/tmp/proj$  scons -v
  SCons by Steven Knight et al.:
script: v3.1.2.bee7caf9defd6e108fc2998a2520ddb36a967691, 2019-12-17 
02:07:09, by bdeegan on octodog
engine: v3.1.2.bee7caf9defd6e108fc2998a2520ddb36a967691, 2019-12-17 
02:07:09, by bdeegan on octodog
engine path: ['/usr/lib/scons/SCons']
  Copyright (c) 2001 - 2019 The SCons Foundation
  

Regards,
robert


-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'unstable'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386


Versions of packages scons depends on:
ii  python33.9.1-1
ii  python3-pkg-resources  51.3.3-1

scons recommends no packages.

scons suggests no packages.

-- no debconf information


scons-test-project.tar.gz
Description: application/gzip

Bug#970210: forcemerge vs. pkgreport.cgi

[Robert Luberda]

severity 970210 important
thanks

Don Armstrong pisze:
> 
> retitle 970210 repeatmerged=no can't be configured to show a bug other than 
> the lowest numbered bug
> severity 970210 wishlist

It's not a wishlist, it's a bug, as the current behavior, that is based
on some strange assumption that a report filled first is always better
than a report filled later, is just against the common sense.
I know that the things have like this for ages, and everybody got used
to it, and got used to having to work-around it by changing the title of
merged reports, but still I consider this as a bug.

> 
> The documentation is pretty clear: "Forcibly merges two or more bug
> reports. The settings of the first bug listed which must be equal in a
> normal merge are assigned to the bugs listed next. 

IMHO nothing here says which bug will be displayed as the "main" bug.

> 
> It doesn't cause a bug to be listed when repeatmerged=no, change the
> title, or anything like that.

Bug report is not only about settings and title, but also its
contents... For example if one bug report contains only information that
something does not work, but the other bug report provides also some
analysis or work-around, than it is obvious to me which of them we
should prefer.
And similarly in the ispell case, https://bugs.debian.org/979575
provides more information than your arbitrarily chosen
https://bugs.debian.org/979549 .

> 
> Hope that clarifies things a bit.

And I hope I convinced you to actually fix this bug.

Regards,
robert

Bug#970210: forcemerge vs. pkgreport.cgi

[Robert Luberda]

severity 970210 important
thanks

On nie, 13 wrz 2020, 積丹尼 Dan Jacobson wrote:

> Package: bugs.debian.org
> 
> One would expect
> forcemerge A B
> to cause
> pkgreport.cgi?package=X
> to list A, not B.
> I mean B gets A's version, etc.

I agree with Dan. I've just forcemerged 979575 979549 979565, and then
marked 979575 as affecting a few packages, for example ifrench.
When I was doing this, I expected 979575 to be the "master" bug, that
will be displayed in bug pages for both ispell and affected packages. 
In other words I thought that ifrench's bugs page would list this bug:

 #979575 ispell 3.4.01 breaks affix files of igerman98 and hkgerman

but it unfortnatelly lists the following one, what seems to have less sense to
me:
 
 #979549 [S|UPu|=☣] [ispell] ingerman: Problems during ispell-autobuild 

Regards,
robert

Bug#979575: ispell 3.4.01 breaks affix files of igerman98 and hkgerman

[Robert Luberda]

reassign 979549 ispell 3.4.01-1
reassign 979565 ispell 3.4.01-1
forcemerge 979575 979549 979565
affects 979575 ingerman iogerman ifrench iesperanto iswiss
tags 979575 pending fixed-upstream
thanks


Roland Rosenfeld pisze:
> 
> In the meantime upstream maintainer released a version 3.4.02 on

Yes, I've noticed it this morning, and it looks like upgrading to that
version fixes the issue.


Regards,
Robert

Bug#977908: mk-configure: Fails with 'Settings for gcc-10.2.1 is not available'

[Robert Luberda]

Package: mk-configure
Version: 0.35.0-1
Severity: serious
Justification: causes FTBFS of libmaa


It looks like mk-configure fails to recognize gcc-10.2.1 (which migrated
to testing almost a week ago), what causes my libmaa package to FTBFS, see:
https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/libmaa.html

   dh "binary" --buildsystem=mkcmake
   dh_update_autotools_config -O--buildsystem=mkcmake
   dh_autoreconf -O--buildsystem=mkcmake
   dh_auto_configure -O--buildsystem=mkcmake
   dh_auto_build -O--buildsystem=mkcmake
mkcmake PREFIX=/usr MANDIR=/usr/share/man INFODIR=/usr/share/info 
SYSCONFDIR=/etc STRIPFLAG= LIBDIR=/usr/lib/x86_64-linux-gnu 
LIBEXECDIR=/usr/lib/x86_64-linux-gnu
==
all ===> doc
==
all ===> maa
checking C compiler type... gcc 10.2.1
bmake[2]: "/usr/share/mk-configure/mk/mkc_imp.platform.sys.mk" line 112: 
'Settings for gcc-10.2.1 is not available, run "mkc_compiler_settings" utility'
*** Error code 1
dh_auto_build: error: mkcmake PREFIX=/usr MANDIR=/usr/share/man 
INFODIR=/usr/share/info SYSCONFDIR=/etc STRIPFLAG= 
LIBDIR=/usr/lib/x86_64-linux-gnu LIBEXECDIR=/usr/lib/x86_64-linux-gnu returned 
exit code 1
make: *** [debian/rules:12: binary] Error 25
dpkg-buildpackage: error: debian/rules binary subprocess returned exit status 2



Regards,
Robert

-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'unstable'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386


Versions of packages mk-configure depends on:
ii  bmake  20200710-5

Versions of packages mk-configure recommends:
ii  pkg-config  0.29.2-1

mk-configure suggests no packages.

-- no debconf information

Bug#977858: apt-listchanges: crashes with name 'syslog' is not defined

[Robert Luberda]

Package: apt-listchanges
Version: 3.22+nmu2
Severity: normal

After setting frontend  to 'syslog' in /etc/apt-listchanges.conf 
(manually, because `dpkg-reconfigure apt-listchanges' does not 
allow me to do it for some reason) apt-listchanges fails with:

  File "/usr/bin/apt-listchanges", line 323, in 
main(config)
  File "/usr/bin/apt-listchanges", line 162, in main
_display(frontend, changes, lambda: _('apt-listchanges: Changelogs'))
  File "/usr/bin/apt-listchanges", line 305, in _display
frontend.display_output(changes)
  File "/usr/share/apt-listchanges/apt_listchanges.py", line 332, in 
display_output
syslog.openlog(facility=syslog.LOG_DAEMON)
NameError: name 'syslog' is not defined


Regards,
robert


-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'unstable'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386


Versions of packages apt-listchanges depends on:
ii  apt2.1.14
ii  debconf [debconf-2.0]  1.5.74
ii  python33.9.0-4
ii  python3-apt2.1.7
ii  python3-debconf1.5.74
ii  sensible-utils 0.0.12+nmu1
ii  ucf3.0043

apt-listchanges recommends no packages.

Versions of packages apt-listchanges suggests:
ii  chromium [www-browser]83.0.4103.116-3.1+b2
ii  elinks [www-browser]  0.13.2-1+b1
ii  firefox-esr [www-browser] 78.6.0esr-1
ii  gnome-terminal [x-terminal-emulator]  3.38.1-2
ii  links [www-browser]   2.21-1
ii  lynx [www-browser]2.9.0dev.6-1
ii  postfix [mail-transport-agent]3.5.6-1
ii  python3-gi3.38.0-1+b2
ii  terminator [x-terminal-emulator]  1.92-2
ii  xterm [x-terminal-emulator]   362-1

-- Configuration Files:
/etc/apt/apt.conf.d/20listchanges changed:
DPkg::Pre-Install-Pkgs { "/usr/bin/apt-listchanges --apt"; };
DPkg::Tools::Options::/usr/bin/apt-listchanges::Version "2";
DPkg::Tools::Options::/usr/bin/apt-listchanges::InfoFD "20";
Dir::Etc::apt-listchanges-main "listchanges.conf";
Dir::Etc::apt-listchanges-parts "listchanges.conf.d";


-- debconf information:
* apt-listchanges/reverse: false
* apt-listchanges/headers: true
* apt-listchanges/confirm: true
* apt-listchanges/no-network: false
* apt-listchanges/save-seen: true
* apt-listchanges/email-format: html
* apt-listchanges/email-address: root
* apt-listchanges/frontend: pager
* apt-listchanges/which: both

Bug#973315: dwww doesnt decompress documentation files

[Robert Luberda]

Calum McConnell writes:

> 
> However, DWWW wanted to download the Gziped readme.md I pointed it to.  Is 
> there
> any way that can be changed, so that it gets decompressed and then served?

The issue is related to files in a markdown format only, so I've
retitled the bug already. I'll add a work-around for it in next dwww
upload to treat markdown files as plain text files

> For instance, .gz extensions could be decompressed,

They are already. However the markdown file was sent as 'text/markdown',
and your browser did not know how to handle it.

> .md fed through kramdown (if > its installed), ect.

I'll work on providing better support next year.

Regards,
Robert

Bug#963680: dwww: Coding error setting $mime_charset

[Robert Luberda]

Howard Johnson writes:
> 
> Starts at line 504 in dwww-convert:
> 
> $mime_charset="UTF-8";

It was $mime_charset="ISO-8859-1" originally, and the code following it
had much more sense than.

Thanks for noticing it.

Regards,
Robert

Bug#946617: dwww: runman prints empty content

[Robert Luberda]

Chad Wallace writes:

Hi,

> Dear Maintainer,
> 
> After upgrading to buster, I failed to get any man pages showing in my
> dwww output.  The  tag is completely empty.
> 
> After editing dwww-convert for debugging, I found an error from man:
> man: nroff: Bad system call
> 
> Then, I found an example of that on another bug report, which led me to
> add a line to dwww-convert:
> 
> ...begin patch
> --- dwww-convert.orig 2019-12-11 14:51:25.131093890 -0800
> +++ dwww-convert  2019-12-11 15:14:02.625790098 -0800
> @@ -246,6 +246,7 @@
>  $dir =~ s/\/[^\/]*$//;
>  chdir ("$dir/..") or die "Can't chdir!\n";
>  $ENV{'MAN_KEEP_FORMATTING'} = 1;
> +$ENV{'MAN_DISABLE_SECCOMP'} = 1;
>  my $IN_FH = ("man  -EUTF-8 -P/bin/cat  -l \"$filename\" 
> 2>/dev/null | dwww-txt2html --man --utf8", "r");
>  chdir ("/");
>  while (<$IN_FH>) {
> ...end patch
> 
> That got it working... but I don't really know why. :-)

I cannot reproduce the issue myself, even after downgrading man-db to
2.8.5-2. The bug is almost a year old, can you still reproduce it yourself?

Anyway I am not sure if setting MAN_DISABLE_SECCOMP inside dwww would be
a good idea, as according to NEWS file in man-db sources, it would hide
issues which man-db wants to be notified about:

* Confine most subprocesses that handle untrusted data using
  seccomp.  This mainly deals with subprocesses that perform
  encoding conversions, (de)compressors, groff programs, and a few
  other odds and ends.  groff programs use a slightly more
  permissive filter since they need to create temporary files, so
  additional path-based confinement (e.g. using AppArmor) is still
  useful.

  If this goes wrong, then MAN_DISABLE_SECCOMP=1 can be set in the
  environment to disable it, but please report any such problem as a
  bug.

I'm adding Collin for confirmation that the above is still valid, and
for a piece of advice what should I do with this bug report.


Regards,
Robert

Bug#977419: lintian: Use of uninitialized value in string eq at package-relations.pm

[Robert Luberda]

Felix Lechner pisze:
Hi,
> 
> On Tue, Dec 15, 2020 at 11:44 AM Francesco Poli
>  wrote:
>>
>> But it seems to work without warnings or errors or false positives!

I can confirm, version from git works for me too, thanks!

> 
> Closing.

I'm not sure why. It is fixed in git, but the current version of lintian
in unstable does not work (i.e. when I downgrade lintian from 2.104+git
to 2.104, without changing any other package, I can still observe the
warnings; when I install 2.104+git the warnings are gone).


Regards,
robert

Bug#977419: lintian: Use of uninitialized value in string eq at package-relations.pm

[Robert Luberda]

On Mon, 14 Dec 2020 22:10:44 -0800 Felix Lechner
 wrote:
> Hi,
> 
> 
> Alas, I *cannot* reproduce it with either of the two packages. I tried
> both 2.104.0 and our development version (on stable). 

I have the same issue for any of my packages, for example:
Warning in group dwww/1.13.5: Use of uninitialized value in string eq at
/usr/share/lintian/checks/fields/package-relations.pm line 129.

I think it is related to the recent perl upgrade. I use the latest
version of perl which is 5.32.0-6.

>Are you both on
> more cutting-edge releases? Specifically, what is your version of
> liblist-moreutils-perl, please? Thanks!

On my system it is 0.430-1

Regards,
robert

Bug#971713: sysstat: init or systemd file has overlapping runlevels

[Robert Luberda]

Trek pisze:

Hi,

> 
> so the fix would be like the one attached to this mail

I've tried this update-rc.d -f sysstat command manually, and then I run
'apt install --reinstall sysstat', and as the result the warning about
default stop level is fixed, but the original warning about overlapping
run-levels is not.


>> insserv: Script ssh has overlapping Default-Start and Default-Stop
>> runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.

What is surprising the ssh warning is shown even if I run insserv from
the command line without giving any other arguments:

robert@vox:/tmp$ sudo insserv
insserv: Script ssh has overlapping Default-Start and Default-Stop
runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.
insserv: FATAL: service mountkernfs has to exist for service udev
insserv: FATAL: service mountdevsubfs has to exist for service hwclock
insserv: FATAL: service urandom has to exist for service networking
insserv: exiting now!

While the sysstat warning is shown only on upgrades.

Anyway I've just checked what the ssh links are:

robert@vox:/tmp$ sudo ls -la /etc/rc*.d/*ssh*
lrwxrwxrwx 1 root root 13 2017-03-27  /etc/rc2.d/K01ssh -> ../init.d/ssh
lrwxrwxrwx 1 root root 13 2017-03-27  /etc/rc3.d/K01ssh -> ../init.d/ssh
lrwxrwxrwx 1 root root 13 2017-03-27  /etc/rc4.d/K01ssh -> ../init.d/ssh
lrwxrwxrwx 1 root root 13 2017-03-27  /etc/rc5.d/K01ssh -> ../init.d/ssh

Removing the links removes the warning as well:

robert@vox:/tmp$ sudo update-rc.d -f ssh remove
insserv: Script ssh has overlapping Default-Start and Default-Stop
runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.

robert@vox:/tmp$ sudo insserv
insserv: FATAL: service mountkernfs has to exist for service udev
insserv: FATAL: service mountdevsubfs has to exist for service hwclock
insserv: FATAL: service urandom has to exist for service networking
insserv: exiting now!

Even apt install --reinstall sysstat does not display any warning.

But when I re-add the K01ssh links, 'apt install --reinstall sysstat'
warns about ssh, and no longer about sysstat service...

Oh, wait, I've just read my email from Sunday, and it looks like the
warning on my system was always about ssh:
>> insserv: warning: current stop runlevel(s) (empty) of script
>>`sysstat' overrides LSB defaults (0 1 6).
>> insserv: Script ssh has overlapping Default-Start and Default-Stop
   ^^^
>> runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.

Sorry about not noticing it before, I must have been too strongly
suggested by the Julian's original bug report :(

So it looks like the warning says that run-level links were customized
by a user. IMHO it would be nice if the warning message stated this
simple fact in a more explicit way.

> 
> this instead comes from ssh, that have an empty Default-Stop too
> 
> @Robert Luberda: I ran out of ideas, your script was correct with an
> empty Default-Stop field, like the ssh one!

I looked into insserv(8) man page before changing the script, and was
under impression that Default-Start and Default-Stop must contain at
least one run level, as the example given at the top of the man page says:

# Default-Start: run_level_1 [ run_level_2 ...]
# Default-Stop:  run_level_1 [ run_level_2 ...]


But if you're sure they can be empty, I will revert the change in
sysstat's init.d script.

Regards,
robert

Bug#749595: doc-base: fails to upgrade

[Robert Luberda]

merge #973797 #749595
thanks

Sami Liedes writes:
> Package: doc-base
> Version: 0.10.5
> Severity: normal
> 
> doc-base fails to upgrade:
> 
> 
> Setting up doc-base (0.10.5) ...
> Processing 7 added doc-base files...
> Cannot open file `/usr/share/doc-base/ocaml-doc-ocaml' for reading: No such 
> file or directory.
> 

Package ocaml-doc in version 4.0.1-1 had two duplicated
files:/usr/share/doc-base/ocaml-doc-ocaml.in
/usr/share/doc-base/ocaml-doc-ocaml

While processing them doc-base displayed the following message:

Error while merging /usr/share/doc-base/ocaml-doc-ocaml with
/usr/share/doc-base/ocaml-doc-ocaml.in: format html already defined.


Then ocaml-doc 4.0.2-1 switched to providing
/usr/share/doc-base/ocaml-doc file only, but doc-base failed to notice
that the other files were removed, what resulted in the following error:
Cannot open file `/usr/share/doc-base/ocaml-doc-ocaml' for reading: No
such file or directory.


Regards,
Robert

Bug#973797: doc-base: package dicomscope-doc is required but not available in testing

[Robert Luberda]

Michael Becker pisze:
> Package: doc-base
> Version: 0.10.9
> Severity: normal
> 
> doc-base can't be installed on testing because it requires file 
> /usr/share/doc-base/dicomscope-doc-dscs from package dicomscope-doc, which is 
> not available on testing
> 

Hi,

It would be helpful if you could be more specific, for example by
including output that you saw on screen. Anyway, I've just managed to
reproduce the bug. I've installed old version of dicomscope-doc (i.e.
3.6.0-20). Now, when trying to upgrade it to the latest version
(3.6.0-22) doc-base fails with:


Setting up doc-base (0.10.9) ...
Processing 2 added doc-base files...
Cannot open file `/usr/share/doc-base/dicomscope-doc-dscs' for reading:
No such file or directory.
dpkg: error processing package doc-base (--configure):
 installed doc-base package post-installation script subprocess returned
error exit status 3
Setting up libsndio6.1:amd64 (1.1.0-3) ...
Processing triggers for libc-bin (2.31-5) ...
Errors were encountered while processing:
 doc-base


I'll fix in next upload. As a work-around, running 'install-docs -I'
command manually helps.

Regards,
robert

Bug#977261: lintian: incorrect program name in --help & --version lines of new commands

[Robert Luberda]

Package: lintian
Version: 2.104.0
Severity: minor


Why trying to find out what has happened with lintian-info -t, I've
found this:

robert@vox:~/TEMP/sysstat/sysstat/debian$ lintian-info -h
Usage: annotate-lintian-hints [log-file...] ...
   annotate-lintian-hints --annotate [overrides ...]
// stripped the rest of output

robert@vox:~$ annotate-lintian-hints
bash: annotate-lintian-hints: command not found

robert@vox:~$ lintian-annotate-hints -h
  ^^^
Usage: annotate-lintian-hints [log-file...] ...
   annotate-lintian-hints --annotate [overrides ...]
   ^
robert@vox:~$ lintian-annotate-hints --version
  ^^^
annotate-lintian-hints v2.104.0
^

Same for the other command:

robert@vox:~$ lintian-explain-tags --help
Usage: explain-lintian-tags [log-file...] ...
   explain-lintian-tags [--tags] tag .


robert@vox:~ lintian-explain-tags  --version
 ^^^
annotate-lintian-hints v2.104.0



Regards,
robert



-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'unstable'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.37-vox (SMP w/4 CPU threads)
Kernel taint flags: TAINT_UNSIGNED_MODULE
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages lintian depends on:
ii  binutils2.35.1-4
ii  bzip2   1.0.8-4
ii  diffstat1.63-1
ii  dpkg1.20.5
ii  dpkg-dev1.20.5
ii  file1:5.39-3
ii  gettext 0.19.8.1-10
ii  gpg 2.2.20-1
ii  intltool-debian 0.35.0+20060710.5
ii  libapt-pkg-perl 0.1.36+b4
ii  libarchive-zip-perl 1.68-1
ii  libcapture-tiny-perl0.48-1
ii  libclass-xsaccessor-perl1.19-3+b6
ii  libclone-perl   0.45-1+b1
ii  libconfig-tiny-perl 2.24-1
ii  libcpanel-json-xs-perl  4.25-1+b1
ii  libdata-dpath-perl  0.58-1
ii  libdata-validate-domain-perl0.10-1
ii  libdevel-size-perl  0.83-1+b2
ii  libdpkg-perl1.20.5
ii  libemail-address-xs-perl1.04-1+b3
ii  libfile-basedir-perl0.08-1
ii  libfile-find-rule-perl  0.34-1
ii  libfont-ttf-perl1.06-1
ii  libhtml-html5-entities-perl 0.004-1
ii  libipc-run3-perl0.048-2
ii  libjson-maybexs-perl1.004003-1
ii  liblist-compare-perl0.55-1
ii  liblist-moreutils-perl  0.416-1+b6
ii  liblist-utilsby-perl0.11-1
ii  libmoo-perl 2.004004-1
ii  libmoox-aliases-perl0.001006-1
ii  libnamespace-clean-perl 0.27-1
ii  libpath-tiny-perl   0.114-1
ii  libperlio-gzip-perl 0.19-1+b7
ii  libproc-processtable-perl   0.59-2+b1
ii  libsereal-decoder-perl  4.018+ds-1+b1
ii  libsereal-encoder-perl  4.018+ds-1+b1
ii  libtext-glob-perl   0.11-1
ii  libtext-levenshteinxs-perl  0.03-4+b8
ii  libtext-markdown-discount-perl  0.12-1+b1
ii  libtext-xslate-perl 3.5.8-1+b1
ii  libtime-duration-perl   1.21-1
ii  libtime-moment-perl 0.44-1+b3
ii  libtimedate-perl2.3300-1
ii  libtry-tiny-perl0.30-1
ii  libtype-tiny-perl   1.012000-1
ii  libunicode-utf8-perl0.62-1+b2
ii  liburi-perl 5.05-1
ii  libxml-libxml-perl  2.0134+dfsg-2+b1
ii  libyaml-libyaml-perl0.82+repack-1+b1
ii  lzip1.21-8
ii  lzop1.04-2
ii  man-db  2.9.3-2
ii  patchutils  0.4.2-1
ii  perl [libdigest-sha-perl]   5.32.0-5
ii  t1utils 1.41-4
ii  unzip   6.0-25
ii  xz-utils5.2.4-1+b1

lintian recommends no packages.

Versions of packages lintian suggests:
pn  binutils-multiarch 
ii  libtext-template-perl  1.59-1

-- no debconf information

Bug#971713: sysstat: init or systemd file has overlapping runlevels

[Robert Luberda]

reassign 971713 insserv 1.21.0-1
severity 971713 grave
affects 971713 sysstat
thanks

Julian Gilbey writes:
> Package: sysstat
> Version: 12.4.0-1
> Severity: normal
> 
> When installing any package that has an init.d file, I get the
> following warning message about sysstat:
> 
> insserv: Script sysstat has overlapping Default-Start and Default-Stop
> runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.


sysstat's init.d has the following lines in /etc/init.d/sysstat

# Default-Start: 2 3 4 5
# Default-Stop:

I'm not sure how empty Default-Stop is interpreted, so I've just tried
to change it to:
# Default-Stop: 0 1 6

As a result insserv displays two warnings instead of one:

insserv: warning: current stop runlevel(s) (empty) of script `sysstat'
overrides LSB defaults (0 1 6).
insserv: Script ssh has overlapping Default-Start and Default-Stop
runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.

As a maintainer of sysstat I have no idea what else I can do to fix the
warning. It seems to me there is some bug in insserv. As the bug affects
my package, and  according to the bug report "any package that has
init.d file", I'm setting severity of this report to grave.

Oh, reinstalling the package does not fix the warnings either:

robert@vox:~/TEMP/sysstat$ grep Default-Stop /etc/init.d/sysstat
# Default-Stop:  0 1 6
robert@vox:~/TEMP/sysstat$ LC_ALL=C sudo dpkg -i
sysstat_12.4.1-1~_amd64.deb
(Reading database ... 315689 files and directories currently installed.)
Preparing to unpack sysstat_12.4.1-1~_amd64.deb ...
Unpacking sysstat (12.4.1-1~) over (12.4.1-1~) ...
Setting up sysstat (12.4.1-1~) ...
insserv: warning: current stop runlevel(s) (empty) of script `sysstat'
overrides LSB defaults (0 1 6).
insserv: Script ssh has overlapping Default-Start and Default-Stop
runlevels (2 3 4 5) and (2 3 4 5). This should be fixed.
robert@vox:~/TEMP/sysstat$ grep Default-Stop /etc/init.d/sysstat
# Default-Stop:  0 1 6
robert@vox:~/TEMP/sysstat$ ls -la /etc/{init.d,rc*}/*sysstat
-rwxr-xr-x 1 root root 1593 12-13 10:03 /etc/init.d/sysstat
lrwxrwxrwx 1 root root   17 12-13 10:01 /etc/rc2.d/S03sysstat ->
../init.d/sysstat
lrwxrwxrwx 1 root root   17 12-13 10:01 /etc/rc3.d/S03sysstat ->
../init.d/sysstat
lrwxrwxrwx 1 root root   17 12-13 10:01 /etc/rc4.d/S03sysstat ->
../init.d/sysstat
lrwxrwxrwx 1 root root   17 12-13 10:01 /etc/rc5.d/S03sysstat ->
../init.d/sysstat

Also please note that sysstat's debian/rules uses the following lines,
but I'm not sure if this information is relevant or not:

override_dh_installinit-arch:
dh_installinit --no-start --no-restart-after-upgrade
--no-stop-on-upgrade

override_dh_installsystemd-arch:
dh_installsystemd --no-start --no-restart-after-upgrade
--no-stop-on-upgrade

BTW. I don't use insserv, I've installed it today, and I've seen these
messages during installation:

Konfigurowanie pakietu insserv (1.21.0-1) ...
insserv: Script ssh has overlapping Default-Start and Default-Stop
runlevels (2 3 4 5) and (2 3 4 5). This should
be fixed.
insserv: FATAL: service mountkernfs has to exist for service udev
insserv: FATAL: service mountdevsubfs has to exist for service hwclock
insserv: FATAL: service urandom has to exist for service networking
insserv: exiting now!

I don't know if they are related or not. I've checked that
/etc/init.d/ssh also contains empty Default-Stop line.


Regards,
Robert

Bug#968416: lintian: Gives ridiculous warning about spelling in overrides file

[Robert Luberda]

Package: lintian
Version: 2.88.0
Severity: normal

While preparing a new upload of the super package, I noticed that 
lintian started to show the following ridiculous warning about 
spelling in a line that is supposed to explain why lintian is 
wrong with its "typo in manual page" warning.

W: super: spelling-in-override-comment typo-in-manual-page (line 3) "allow to" 
"allow one to"


Here is the full content of lintian-overrides:

super: setuid-binary usr/bin/super 4755 root/root
# "...changes root's default setting from default-allow to default-deny..."
super: spelling-error-in-manpage usr/share/man/man5/super.tab.5.gz allow to 
allow one to


Please note that the warning is talking about "(line 3)" while the
actual problem lies in line 2. I've lost some time to discover this.

But to be honest. I have no idea, why lintian even checks spelling of the
overrides files. The files are not supposed to be read by regular users,
so what's the point of checking spelling there? And BTW. are you going
to check spelling of comments in source code files as well? I'm pretty
sure you can find a lot of spelling typos there...


Regards,
robert


Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Bug#954743: lintian: orig-tarball-missing-upstream-signature does not say how to add it to uploaded tarballs

[Robert Luberda]

Chris Lamb writes:

Hi

>
> In other words, my "AIUI" should be replaced with something much
> stronger; I've done many "useless" sourceful uploads successfully, but
> I've never attempted to add an .sig…

It looks like upload with '-sa' worked, at least for experimental:

https://tracker.debian.org/news/209/accepted-wmnd-0417-4exp2-source-into-experimental/

The asc file was stored inside pool directory:

https://deb.debian.org/debian/pool/main/w/wmnd/wmnd_0.4.17.orig.tar.gz.asc

so I was also able to make an upload without '-sa' afterwards:

https://tracker.debian.org/news/213/accepted-wmnd-0417-4exp3-source-into-experimental/

So I think it will be OK to do upload for unstable, however I still need
to first check how to store the signature in my git repository. I did
some tests few days ago, but I haven't got time to finish it. I'll try
to do this before this weekend.

>
>> Honestly, I'm somewhat partial on putting such archive-dependant
>> information in lintian is alright (dak is only one of the many debian
>> archive software out there…).

Yes, but IMHO it would be nice if for example there was some wiki page
online with all the information, and lintian had a link to it.

>
> (I don't believe that is being requested here.)

I think lintian makes a really good job in improving quality of Debian
packages, and I treat all the tags it gives pretty seriously, especially
if they are warnings or errors. However I would like to have a clear
straightforward information how to deal with those tags. Vast majority
of tags' descriptions is clear, but
'orig-tarball-missing-upstream-signature' raised my concerns, so I
decided to create this bug report.

Regards,
robert

Bug#954743: lintian: orig-tarball-missing-upstream-signature does not say how to add it to uploaded tarballs

[Robert Luberda]

Chris Lamb pisze:
> Hi Robert,

Hi,

> 
>> The upstream tarball was uploaded to Debian archive in 2013, and there
>> have been no new upstream releases since then. 
> 
> Did you try the recently-closed #872864 references in that tag's long
> description, just out of interest?


Yes, but I don't think it contains answer to my concerns.

I've just put the *.sig file in the directory where the package is built
from, what made lintian to stop complaining.

But I am still not sure if I am allowed to upload a package in which the
*.dsc file refers to a *.asc file, without uploading the *.asc file itself.
Should use the '-sa' option to force both the *.orig.tar.gz and *.asc
into .*.changes? I think that such an upload will be rejected by DAK, as
the *.tar.gz is already in the archive. But to be honest this is based
on my past experiences, but maybe something has changed since then. Is
re-uploading upstream tarballs permitted now?

Regards,
robert

Bug#954743: lintian: orig-tarball-missing-upstream-signature does not say how to add it to uploaded tarballs

[Robert Luberda]

Package: lintian
Version: 2.58.0
Severity: normal

lintian started to complain that wmnd source tarball lacks upstream
signature:

W: wmnd source: orig-tarball-missing-upstream-signature wmnd_0.4.17.orig.tar.gz

The upstream tarball was uploaded to Debian archive in 2013, and there
have been no new upstream releases since then. Please explain in the
description of the orig-tarball-missing-upstream-signature tag, how to
add this signature to a new upload: Should I use 'dpkg-genchanges -sa'
in this case, and won't DAK reject such upload then? (Not to mention
that I have no idea how to convince git-buildpackage to store the
signature in git repo for the existing tarball).

Regards,
robert

-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Bug#953055: apt-listchanges: DeprecationWarning: PY_SSIZE_T_CLEAN [...] after python3.7->python3.8

[Robert Luberda]

reassign 953055 python3-apt 1.8.6
affects 953055 apt-listchanges
thanks

Anders Breindahl writes:
> 
> Dear Maintainer,
> 
> As unattended-upgrades updated python (between 06:27:18 and 06:28:21),
> the output behaviour of apt-listchanges changed in the subsequent run
> (starting 06:28:21):
> 
> Log started: 2020-03-03  06:28:21
> /usr/bin/apt-listchanges:209: DeprecationWarning: PY_SSIZE_T_CLEAN will 
> be required for '#' formats
>   if fromversion and apt_pkg.version_compare(fromversion, version) >= 0:


I've just done a few tests, using a command like the following:
  /usr/bin/apt-listchanges -f text --since 9: any_package.deb
and I can reproduce the warning when python3-apt 1.8.6 is installed, but
the warning is gone after upgrading python3-apt to 1.9.9.

It looks like the bug can be closed now; however I am reassigning it,
because it might be better if python3-apt maintainer confirms it,
especially that I have no idea what the PY_SSIZE_T_CLEAN warning is
actually about.

Thanks,
robert

Bug#951081: git-gui: "missing close-bracket" error while doing branch->reset

[Robert Luberda]

Package: git-gui
Version: 1:2.25.0-1
Severity: normal

After selecting Branch->Reset... and then answering 'Yes' to the
question, a popup with the following error appears:


  missing close-bracket
  missing close-bracket
  while executing
  "set status_bar_operation ["
  invoked from within
  "if {[ask_popup $op_question] eq {yes}} {
set fd [git_read --stderr read-tree --reset -u -v HEAD]
fconfigure $fd -blocking 0 -translation binary
..."
  (procedure "merge::reset_hard" line 28)
  invoked from within
  "merge::reset_hard"
  invoked from within
  ".#mbar.#mbar#branch invoke active"
  ("uplevel" body line 1)
  invoked from within
  "uplevel #0 [list $w invoke active]"
  (procedure "tk::MenuInvoke" line 50)
  invoked from within
  "tk::MenuInvoke .#mbar.#mbar#branch 1"
  (command bound to event)


Regards,
robert

-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable'), (500, 'stable-updates'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386


Versions of packages git-gui depends on:
ii  git  1:2.25.0-1
ii  tk   8.6.9+1+b1

Versions of packages git-gui recommends:
ii  gitk  1:2.25.0-1

Versions of packages git-gui suggests:
ii  aspell   0.60.8-1
pn  git-doc  
pn  meld 

-- no debconf information

Bug#946657: CVE-2019-19725 not fixed in 12.2.0-1

[Robert Luberda]

Salvatore Bonaccorso writes:
> Control: reopen -1
> Control: found -1 12.2.0-1
>
Hi,


>>  sysstat (12.2.0-1) unstable; urgency=medium
>>  .
>>* New upstream stable version:
>>  + fixes double free in check_file_actlst in check_file_actlst in
>>sa_common.c (CVE-2019-19725, closes: #946657).
> 
> But this is not actually true I believe.
> https://github.com/sysstat/sysstat/commit/a5c8abd4a481ee6e27a3acf00e6d9b0f023e20ed
> is not applied in 12.2.0-1, and I do not see it applied as patch as

I don't know why, but I've assumed that 12.2.0 fixed the issue :(
Thanks for noticing my mistake; I'll apply the upstream patch in -2 shortly.

Regards,
robert

Bug#947115: lintian: Bogus no-dh-sequencer

[Robert Luberda]

Package: lintian
Version: 2.42.0
Severity: minor

While preparing new upload for sysstat package I've noticed this:

  I: sysstat source: no-dh-sequencer

I've checked what that means:

  lintian-info --tag no-dh-sequencer
  I: no-dh-sequencer
  N:
  N:   The debian/rules does not use the dh &@ sequencer.

I have no idea what is "dh &@ sequencer", but sysstat's
debian/rules contain this:

  %:
dh ${@}


Regards,
robert


-- System Information:
Debian Release: bullseye/sid
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Versions of packages lintian depends on:
ii  binutils 2.33.1-6
ii  bzip21.0.8-2
ii  diffstat 1.63-1
ii  dpkg 1.19.7
ii  dpkg-dev 1.19.7
ii  file 1:5.37-6
ii  gettext  0.19.8.1-10
ii  gpg  2.2.17-3
ii  intltool-debian  0.35.0+20060710.5
ii  libapt-pkg-perl  0.1.36+b2
ii  libarchive-zip-perl  1.67-1
ii  libberkeleydb-perl   0.62-1+b1
ii  libcapture-tiny-perl 0.48-1
ii  libcgi-pm-perl   4.44-1
ii  libclass-accessor-perl   0.51-1
ii  libclass-xsaccessor-perl 1.19-3+b3
ii  libclone-perl0.43-2
ii  libdpkg-perl 1.19.7
ii  libemail-valid-perl  1.202-1
ii  libfile-basedir-perl 0.08-1
ii  libfile-find-rule-perl   0.34-1
ii  libfont-ttf-perl 1.06-1
ii  libio-async-loop-epoll-perl  0.20-1
ii  libio-async-perl 0.75-1
ii  libipc-run-perl  20180523.0-2
ii  liblist-compare-perl 0.53-1
ii  liblist-moreutils-perl   0.416-1+b5
ii  libmldbm-perl2.05-2
ii  libmoo-perl  2.003006-1
ii  libmoox-aliases-perl 0.001006-1
ii  libnamespace-clean-perl  0.27-1
ii  libpath-tiny-perl0.108-1
ii  libtext-levenshtein-perl 0.13-1
ii  libtimedate-perl 2.3000-2
ii  libtry-tiny-perl 0.30-1
ii  libtype-tiny-perl1.008000-1
ii  liburi-perl  1.76-1
ii  libxml-libxml-perl   2.0134+dfsg-1+b1
ii  libyaml-libyaml-perl 0.80+repack-2+b1
ii  man-db   2.9.0-2
ii  patchutils   0.3.4-2+b1
ii  perl [libdigest-sha-perl]5.30.0-9
ii  t1utils  1.41-3
ii  xz-utils 5.2.4-1+b1

Versions of packages lintian recommends:
pn  libperlio-gzip-perl  

Versions of packages lintian suggests:
pn  binutils-multiarch 
ii  libhtml-parser-perl3.72-3+b4
ii  libtext-template-perl  1.58-1

-- no debconf information

Bug#908370: afterstep FTCBFS: uses AC_TRY_RUN

[Robert Luberda]

Helmut Grohne writes:

Hi,

> It still fails building manual pages
> with an Exec format error. If the manual pages were moved to
> afterstep-data, one could maybe skip the manual page generation step.

To be honest I am not sure how to fix it. I could move afterstep's
manpages to afterstep-data, but there are still some manpages in
libafterimage-dev - those manpages shouldn't be moved to afterstep-data.

> Can you still apply the attached patch and close this bug when doing so?

Yes, I'll do it soon.

Sorry for late response.

Regards,
robert

Bug#944525: dpkg: Should restore root's PATH while spawning shell for conffile handling

[Robert Luberda]

Package: dpkg
Version: 1.19.7
Severity: normal


I noticed some time ago that my pretty old custom dpkg-conffile-diff 
script is not available (without giving full path to it, which is
/usr/lcoal/sbin/dpkg-conffile-diff) in the shell spawned by dpkg to 
handle config file merges during installations.

Today I've found some time to investigate it, and found this commit 
in apt:
  
https://salsa.debian.org/apt-team/apt/commit/806e94dcd8dbdf7bf1909657fd4331cfe17b4ab0
Set PATH=/usr/sbin:/usr/bin:/sbin:/bin when running dpkg

This avoids a lot of problems from local installations of
scripting languages and other stuff in /usr/local for which
maintainer scripts are not prepared.

(To be honest I am not sure if such change should be done in apt;
if it is really needed then most probably dpkg should do it on its own).

It would be nice if dpkg could restore the default root path, i.e.
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
for a shell spawned for interactive config file handling.

If you think this bug should be fixed by apt somehow, feel free to
reassign it.


Regards,
robert


-- Package-specific info:

-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Versions of packages dpkg depends on:
ii  libbz2-1.0   1.0.8-2
ii  libc62.29-3
ii  liblzma5 5.2.4-1+b1
ii  libselinux1  2.9-2+b2
ii  tar  1.30+dfsg-6+b1
ii  zlib1g   1:1.2.11.dfsg-1+b1

dpkg recommends no packages.

Versions of packages dpkg suggests:
ii  apt1.8.4
pn  debsig-verify  

-- no debconf information

Bug#944083: libbsd-dev: Lacks user_from_uid() & uid_from_user() functions

[Robert Luberda]

Package: libbsd-dev
Version: 0.10.0-1
Severity: wishlist

Hi,

I'm trying to prepare new version of bsd-mailx package from the OpenBSD
repository, but it unfortunatelly fails to compile because they switched
to using to user_from_uid() and uid_from_user(), see the links below.
 Would it be possible to provide the functions in the libbsd package?



https://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin/mail/v7.local.c?rev=1.18=text/x-cvsweb-markup=date
https://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin/mail/v7.local.c.diff?r1=1.17=1.18=date=h
https://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin/mail/temp.c.diff?r1=1.17=1.18=date=h

Regards,
robert



-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')

Versions of packages libbsd-dev depends on:
ii  libbsd0  0.10.0-1

libbsd-dev recommends no packages.

libbsd-dev suggests no packages.

-- no debconf information

Bug#925947: apt-listchanges: warns on postinst: -V is ignored in pypycompile

[Robert Luberda]

reassign 925947 dh-python 4.20191017
thanks

Julian Gilbey writes:

Hi,
> Package: apt-listchanges
> Version: 3.19
> Severity: normal
> 
> On installing, I noticed the following warning:
> 
> Setting up apt-listchanges (3.19) ...
> -V is ignored in pypycompile


This warning was fixed by a chance in apt-listchanges 3.20 due to
removal of X-Python3-Version field from debian/control.

However if there is a need to re-add the field in the future the warning
will re-appear. That's why I'm reassigning to dh-python, as I think it
should strip the -V flag from calls to pypycompile or pypy3compile.
(Alternatively the bug could be re-assigned to pypy/pypy3 packages not
to generate the warning when -V is used)

Regards,
robert

Bug#934937: man-db: Refers to undocumented(7) which has gone

[Robert Luberda]

Package: man-db
Version: 2.8.6.1-1
Severity: minor

The undocumented(7) man page was removed 3 years ago:

  manpages (4.04-2) unstable; urgency=medium
  
... strip other changelog entries ...

* Remove obsolete manpage undocumented.7.
  This manpage was provided in order to comply with an ancient
  policy version. The manpage is no longer needed since Nov 2002.
  Closes: #506653, LP:#426359
  
   -- Dr. Tobias Quathamer   Mon, 15 Feb 2016 14:02:08 +0100


but man-db stil refers to it:

  robert@vox:/tmp$ /usr/bin/man -L C 8 test
  No manual entry for test in section 8
  See 'man 7 undocumented' for help when manual pages are not available.



Regards,
robert

PS. It looks like  ancient bug #32019 ('man-db: Move undocumented(7) from 
manpages to
 man-db') might be no longer relevant.


-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.37-vox (SMP w/4 CPU cores)
Kernel taint flags: TAINT_CRAP, TAINT_UNSIGNED_MODULE
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages man-db depends on:
ii  bsdmainutils   11.1.2+b1
ii  debconf [debconf-2.0]  1.5.73
ii  dpkg   1.19.7
ii  groff-base 1.22.4-3
ii  libc6  2.28-10
ii  libgdbm6   1.18.1-4+b1
ii  libpipeline1   1.5.1-2
ii  libseccomp22.4.1-2
ii  zlib1g 1:1.2.11.dfsg-1+b1

man-db recommends no packages.

Versions of packages man-db suggests:
pn  apparmor   
ii  chromium [www-browser] 76.0.3809.100-1
ii  elinks [www-browser]   0.13~20190125-3
ii  firefox-esr [www-browser]  60.8.0esr-1
ii  groff  1.22.4-3
ii  less   487-0.1+b1
ii  links [www-browser]2.19-2
ii  lynx [www-browser] 2.9.0dev.2-1

-- debconf information excluded

Bug#869733: ispell.1X: Fix some formatting and textual matters

[Robert Luberda]

Bjarni Ingi Gislason writes:

Hi,

> Package: ispell
> Version: 3.4.00-6
> Severity: minor
> 
>   The patch is in the attachment.

Sorry for the delay, Thanks for the patch, I've just applied it in git,
see [1].

As your patch was apparently for the older version of ispell (3.3.02),
some of your corrections were already there. Additionally I've changed
your fix for:
  "-munching a normal-sized dictionary (15K roots, 45K expanded words)"
from:
  "+munching a normal-sized dictionary (15\ kB roots, 45\ kB expanded
words)"
to:
  "+munching a normal-sized dictionary (15000 roots, 45000 expanded words)"
as K meaning just "kilo", i.e. "thousand", here seemed more sensible to me.

[1]
https://salsa.debian.org/debian/ispell/blob/7416a731b3efb8586bca56099194e2fc66996399/debian/patches/0039-Man-page-issues-fix.patch

Regards,
robert

Bug#934191: ftp.debian.org: Should reject/ignore/auto-binNMU binary uploads

[Robert Luberda]

Package: ftp.debian.org
Severity: serious
Justification: common sense

As the binary uploads are usually pretty much useless, because they cannot
migrate to testing anymore [1], ftp.debian.org should do one of the
following things:
 - reject them (what seems the least preferred option to me);
 - ignore them (i.e. accept upload, but copy all included *.deb files to 
/dev/null);
 - automatically schedule binNMU;
 - or anything other that won't force developers to waste their time on 
   preparing another upload with no real changes, only to have their packages
   migrated to testing.

Regards,
robert

[1] https://lists.debian.org/debian-devel-announce/2019/07/msg2.html

Bug#933869: O: smb2www -- SMB/CIFS network client with a web interface

[Robert Luberda]

Package: wnpp
Severity: normal

I am orphaning the smb2www package, as I do not use it any longer.

I've just made an upload with sets maintainer to Debian QA:

Package: smb2www
Architecture: all
Version: 980804-42
Priority: optional
Section: net
Maintainer: Debian QA Group 
Installed-Size: 216
Depends: apache2 | httpd-cgi, smbclient, debconf (>= 0.5) | debconf-2.0, perl
Recommends: apache2 (>= 2.4.6-4~) | httpd
Filename: release/smb2www_980804-42_all.deb
Size: 92692
MD5sum: 8a5422a78b37a91a602b799b3a37a846
SHA1: 3ccb8eee5a3d5666b48212c7e900b57d5067e20c
SHA256: 91cf36db3c8303f4629b4e96d1aea879c93c28270269f49740744b8e323d0872
SHA512: 
73f19a23a53786e53c1619bbd6e6c9f63073b717968a5073ae9b5e239aae139d19fb13287d80509256e423d46b8b11e91953c37c1d44647ae5c57dd392b11317
Description-en: SMB/CIFS network client with a web interface
 This package allows browsing an SMB/CIFS-based network (such as
 a Windows-based network) by using a
 standard web browser. It uses the Samba smbclient utility.
Description-md5: c47a0f0117037ca5379d08b23c008f6d


signature.asc
Description: PGP signature

Bug#933576: debhelper: dh_installman is terribly slow

[Robert Luberda]

Package: debhelper
Version: 12.2.3
Severity: normal

Hi,

While working on my manpages-pl source package, I've noticed that the
dh_installman step takes more time to execute than all other build steps 
together.

This poor performance is caused by recoding all (i.e. about 1500 in case
of manpages-pl) man pages into UTF-8, what is pretty much useless in case 
of my package, because the pages are in UTF-8 already. 

It would be nice if dh_installman could have some option to disable recoding 
or if it could at least filter the manpages to recode with `isutf8 -l'
or similar command. (I've just checked that 
'isutf8 -l debian/tmp/usr/share/man/pl/man*/*' inside the package is really
quick to determine that all files are in UTF-8).

Regards,
robert

-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.37-vox (SMP w/4 CPU cores)
Kernel taint flags: TAINT_UNSIGNED_MODULE
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages debhelper depends on:
ii  autotools-dev20180224.1
ii  dh-autoreconf19
ii  dh-strip-nondeterminism  1.4.0-1
ii  dpkg 1.19.7
ii  dpkg-dev 1.19.7
ii  dwz  0.12.20190723-1
ii  file 1:5.37-5
ii  libdpkg-perl 1.19.7
ii  man-db   2.8.5-2
ii  perl 5.28.1-6
ii  po-debconf   1.0.21

debhelper recommends no packages.

Versions of packages debhelper suggests:
pn  dh-make  

-- no debconf information

Bug#925185: unblock pre-approval: sysstat/12.0.3-1 (actually 12.0.3-2)

[Robert Luberda]

Control: tags -1 -moreinfo
Control: retitle -1  unblock sysstat/12.0.3-2

Hi,

>> [...]
> Please go ahead with 12.0.3-1 for buster and remove the moreinfo tag
> when it is ready for unblocks.
> 

Thanks. I uploaded 12.0.3-2 to unstable yesterday. Could you please
unblock it?

Comparing to 12.0.3-1, it only introduces a new changelog entry, and a
fix for the typo you've found:

-- sysstat-12.0.3/debian/changelog 2019-03-17 23:09:46.0 +0100
+++ sysstat-12.0.3/debian/changelog 2019-04-06 09:18:26.0 +0200
@@ -1,3 +1,9 @@
+sysstat (12.0.3-2) unstable; urgency=medium
+
+  * Upload to unstable.
+
+ -- Robert Luberda   Sat, 06 Apr 2019 09:18:26 +0200
+
 sysstat (12.0.3-1) experimental; urgency=medium

   * New upstream stable version:
@@ -11,7 +17,7 @@
 marker into statistics file on systems on which systemd is not used.
 Thanks to Georgios Zarkadas for noticing this (closes: #924864).
   * debian/rules: replace deprecated dh_systemd_start by dh_installsystemd,
-as suggested by lintian; the former command wass ignored by
debhelper v11,
+as suggested by lintian; the former command was ignored by
debhelper v11,
 what in turn resulted in the `--no-start' option being ignored, and the
 restart markers were incorrectly added during package upgrades.


Regards,
robert

Bug#925498: afterstep: v2.2.12-12 FTBFS due to patch#57 generates libAfterImage/Makefile incorrectly.

[Robert Luberda]

Jim Turner pisze:

Hi,

> truetype fonts, and instead use pkg-config.  Problem is that the patch
> properly removes it from configure.in, but NOT from configure itself

The configure script is regenerated during the build process, so there
is no need to change it via a patch.

> I simply untacked the latest source package v2.2.12-12 from Debian Testing
> (http://ftp.gr.debian.org/debian/) on March 25, 2019, ran:
> 
> ./configure
> make


This is not a proper way to build Debian packages. Use 'quilt push -a &&
debian/rules binary' or tools like dpkg-buildpackage or debuild instead.


Regards,
robert

Bug#925187: unblock: apt-listchanges/3.19

[Robert Luberda]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Please unblock package apt-listchanges. The version constains debconf
translation updates only.

unblock apt-listchanges/3.19


Debdiff:

Comparing apt-listchanges_3.18.dsc apt-listchanges_3.19.dsc
diff -Nru apt-listchanges-3.18/debian/changelog 
apt-listchanges-3.19/debian/changelog
--- apt-listchanges-3.18/debian/changelog   2018-11-29 11:37:28.0 
+0100
+++ apt-listchanges-3.19/debian/changelog   2019-03-17 23:48:06.0 
+0100
@@ -1,3 +1,11 @@
+apt-listchanges (3.19) unstable; urgency=medium
+
+  * Updated debconf translations:
++ Danish (Joe Dalton; closes: #923063);
++ Russian (Lev Lamberov, closes: #920599).
+
+ -- Robert Luberda   Sun, 17 Mar 2019 23:48:06 +0100
+
 apt-listchanges (3.18) unstable; urgency=medium
 
   [ Matt Kraai ]
diff -Nru apt-listchanges-3.18/debian/po/da.po 
apt-listchanges-3.19/debian/po/da.po
--- apt-listchanges-3.18/debian/po/da.po2018-11-29 11:37:28.0 
+0100
+++ apt-listchanges-3.19/debian/po/da.po2019-03-17 23:48:06.0 
+0100
@@ -1,15 +1,15 @@
 # Danish translation for apt-listchanges.
-# Copyright (C) 2011 apt-listchanges og nedenstående oversættere.
+# Copyright (C) 2019 apt-listchanges og nedenstående oversættere.
 # This file is distributed under the same license as the apt-listchanges 
package.
 # Claus Hindsgaul , 2004, 2006.
-# Joe Hansen (joedalt...@yahoo.dk), 2011.
+# Joe Hansen (joedalt...@yahoo.dk), 2011, 2019.
 #
 msgid ""
 msgstr ""
-"Project-Id-Version: da\n"
+"Project-Id-Version: apt-listchanges\n"
 "Report-Msgid-Bugs-To: apt-listchan...@packages.debian.org\n"
 "POT-Creation-Date: 2017-11-12 23:48+0100\n"
-"PO-Revision-Date: 2011-01-10 19:25+0200\n"
+"PO-Revision-Date: 2019-02-23 19:25+0200\n"
 "Last-Translator: Joe Hansen \n"
 "Language-Team: Danish \n"
 "Language: da\n"
@@ -145,23 +145,23 @@
 #. Choices
 #: ../templates:4001
 msgid "html"
-msgstr ""
+msgstr "html"
 
 #. Type: select
 #. Description
 #: ../templates:4002
 msgid "Format of e-mail messages:"
-msgstr ""
+msgstr "Format på e-postbeskeder:"
 
 #. Type: select
 #. Description
 #: ../templates:4002
-#, fuzzy
-#| msgid "Please choose which type of changes should be displayed with APT."
 msgid ""
 "Please choose a format for e-mail copies of the displayed changes - either "
 "plain text or HTML with clickable links."
-msgstr "Vælg Hvilke typer ændringer, der skal vises med APT."
+msgstr ""
+"Vælg et format for e-postkopier over viste ændringer - enten ren tekst eller "
+"HTML med henvisninger der kan følges."
 
 #. Type: boolean
 #. Description
@@ -195,11 +195,8 @@
 #. Type: boolean
 #. Description
 #: ../templates:6001
-#, fuzzy
-#| msgid "Should apt-listchanges skip changes that have already been seen?"
 msgid "Insert headers before changelogs?"
-msgstr ""
-"Skal apt-listchanges springe de ændringer over, der er blevet vist før?"
+msgstr "Indsæt teksthoveder før ændringslogge?"
 
 #. Type: boolean
 #. Description
@@ -209,6 +206,9 @@
 "the name of the package, and the names of the binary packages which are "
 "being upgraded (when different from the source package name)."
 msgstr ""
+"apt-listchanges kan indsætte et teksthoved før hver pakkes ændringslog, "
+"der viser pakkenavnet og navnene på de binære pakker, der bliver opgraderet "
+"(når forskellig fra kildepakkenavnet)."
 
 #. Type: boolean
 #. Description
@@ -217,12 +217,14 @@
 "Note however that displaying headers might make the output a bit harder to "
 "read as they might contain long lists of names of binary packages."
 msgstr ""
+"Bemærk dog at visning af teksthoveder kan gøre resultatet en smule sværere "
+"at læse, da de kan indeholde lange lister med navne på binære pakker."
 
 #. Type: boolean
 #. Description
 #: ../templates:7001
 msgid "Disable retrieving changes over network?"
-msgstr ""
+msgstr "Deaktiver indsamling af ændringer over netværk?"
 
 #. Type: boolean
 #. Description
@@ -232,6 +234,9 @@
 "listchanges by default executes the command \"apt-get changelog\", which "
 "tries to download changelog entries from the network."
 msgstr ""
+"I sjældne tilfælde når en binær pakke ikke indeholder en ændringslog-fil, "
+"kører apt-listchanges som stndard kommandoen »apt-get changelog«, som "
+"forsøger at hente ændringslog-poster fra netværket."
 
 #. Type: boolean
 #. Description
@@ -240,12 +245,14 @@
 "This option can disable this behavior, which might for 

Bug#925185: unblock pre-approval: sysstat/12.0.3-1 (actually 12.0.3-2)

[Robert Luberda]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Please approve sysstat 12.0.3, which is upstream bugfix release,
for uploading to unstable and migrating to testing.

The upstream release contains fix for CVE-2018-19416 [1] and 
CVE-2018-19517 [2]; however the patch [3] is not easily applicable
to the version in buster (12.0.1-1), because it depends on another 
patch [4], which contains a fix for a backward compatibility issue
introduced in 12.0.1.  Apart from the two quite a big patches, the
new upstream a few smaller fixes, like the one related to a fix
for infinite loop [5]. In my opinion it should be quite safe to 
allow it for buster, most probably safer than trying to backport 
the patch [3] to 12.0.1 with getting rid of dependency on [4].

The debian packaging part contains fixes for two small regressions 
against current stretch version of sysstat: one is for init
script failure when systemd is not used [6], and another one is for 
unnecessary execution of systemd service file during upgrades.

[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=914384
[2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=914553
[3] 
https://github.com/sysstat/sysstat/commit/bf203d645110ecba8ec3a37874b577ce40a2788b
[4] 
https://github.com/sysstat/sysstat/commit/87bce40bc02ff77edee44a7b9d8233ae6a056012
[5] 
https://github.com/sysstat/sysstat/commit/45de3c27697d9c1c4d8feb12c865d1fe53ce45bf
[6] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=924864

I uploaded systat 12.0.3-1 to experimental a few days ago with the
following changelog:

  sysstat (12.0.3-1) experimental; urgency=medium
  
* New upstream stable version:
  + sadf: Fix out of bound reads security issues (CVE-2018-19416 and
CVE-2018-19517, closes: #914384, #914553);
  + sadf: Fix possible infinite loop;
  + sar: Fortify remap_struct() function to prevent possible crashes on
reading binary datafiles generated by older versions of sysstat.
* systat.init.d: revert a change introduced in 11.5.5-1, as it caused
  the start script to fail to execute the command that adds "Linux Restart"
  marker into statistics file on systems on which systemd is not used.
  Thanks to Georgios Zarkadas for noticing this (closes: #924864).
* debian/rules: replace deprecated dh_systemd_start by dh_installsystemd,
  as suggested by lintian; the former command wass ignored by debhelper v11,
  what in turn resulted in the `--no-start' option being ignored, and the
  restart markers were incorrectly added during package upgrades.
  
   -- Robert Luberda   Sun, 17 Mar 2019 23:09:46 +0100

The debdiff against buster is attached. 

If you think this version would be OK for buster, then I can upload -2
to unstable, with no other changes, except for Debian changelog entry.

Otherwise please let me know what would you approve, and what I should do:
 - backport patch [3] only (but I don't think this would be safer);
 - backport both patches, i.e. [3], and [4] (but those are the biggest ones);
 - something else.

Regards,
robert


-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.0-2-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_OOT_MODULE
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)


sysstat_12.0.3-1.diff.gz
Description: application/gzip


signature.asc
Description: PGP signature

Bug#924867: w-scan: Does not scan for Polish MUX-8 (on VHF) channels

[Robert Luberda]

Package: w-scan
Version: 20170107-2
Severity: normal

According to a comment in w-scan source "VHF no longer used in Europe" [1].
However this is not true for Poland, which used VHF for channels in so called 
MUX-8, see [2].

I've checked today that simply uncommenting the line [1] and then
re-compiling w-scan did not work: even though such a w-scan 
seemed to scan 7MHz frequencies, its output did not contain 
any MUX-8 channel.

However when I downgraded w-scan to version 20130331-1 (using the deb
package from [3]), the `w_scan -c PL -X -f3 -E0 -R0' command was able to 
find all the MUX-8 channels:
   TVP 
Rozrywka(EmiTel):19150:INVERSION_AUTO:BANDWIDTH_7_MHZ:FEC_AUTO:FEC_AUTO:QAM_AUTO:TRANSMISSION_MODE_AUTO:GUARD_INTERVAL_AUTO:HIERARCHY_AUTO:8022:8028:34
   TVP Sport 
HD(EmiTel):19150:INVERSION_AUTO:BANDWIDTH_7_MHZ:FEC_AUTO:FEC_AUTO:QAM_AUTO:TRANSMISSION_MODE_AUTO:GUARD_INTERVAL_AUTO:HIERARCHY_AUTO:8042:8043:54
   
Metro(EmiTel):19150:INVERSION_AUTO:BANDWIDTH_7_MHZ:FEC_AUTO:FEC_AUTO:QAM_AUTO:TRANSMISSION_MODE_AUTO:GUARD_INTERVAL_AUTO:HIERARCHY_AUTO:8062:8063:57
   Zoom 
TV(EmiTel):19150:INVERSION_AUTO:BANDWIDTH_7_MHZ:FEC_AUTO:FEC_AUTO:QAM_AUTO:TRANSMISSION_MODE_AUTO:GUARD_INTERVAL_AUTO:HIERARCHY_AUTO:8082:8084:58
   Nowa 
TV(EmiTel):19150:INVERSION_AUTO:BANDWIDTH_7_MHZ:FEC_AUTO:FEC_AUTO:QAM_AUTO:TRANSMISSION_MODE_AUTO:GUARD_INTERVAL_AUTO:HIERARCHY_AUTO:8102:8108:59
   
WP(EmiTel):19150:INVERSION_AUTO:BANDWIDTH_7_MHZ:FEC_AUTO:FEC_AUTO:QAM_AUTO:TRANSMISSION_MODE_AUTO:GUARD_INTERVAL_AUTO:HIERARCHY_AUTO:8122:8124:60


[1] https://sources.debian.org/src/w-scan/20170107-2/countries.c/#L348
[2] 
https://www.emitel.pl/en/strefa-klienta-2/odbior-rtv/jak-poprawnie-odbierac-mux-8/
[3] 
http://snapshot.debian.org/archive/debian/20130512T154713Z/pool/main/w/w-scan/w-scan_20130331-1_amd64.deb

Regards,
robert


-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.0-2-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_OOT_MODULE
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages w-scan depends on:
ii  libc6  2.28-8

w-scan recommends no packages.

w-scan suggests no packages.

-- no debconf information

Bug#748631: apt-listchanges: doens't list changes for all packages

[Robert Luberda]

Christoph Anton Mitterer writes:

Hi,

> 
> (better 4 years late than never?! * ashamed *)
> 
> It's really the case for all versions of at least and util-linux / lvm2
> packages.
> 
> I'm running sid, so take any version of the these packages from the
> last years and it wouldn't show me the changelogs for them.


It works for me. I've just checked recent mails from apt-listchanges,
and I've found entries for util-linux, for example:

  --- Changes for util-linux (bsdutils libuuid1 libblkid1 libfdisk1
libsmartcols1 libmount1 fdisk util-linux mount uuid-dev
util-linux-locales) ---
  util-linux (2.32.1-0.2) unstable; urgency=medium

* Non-maintainer upload.

[ Niels Thykier ]
* Declare the explicit requirement for (fake)root
[ strip most of the log - RL ]
   -- Andreas Henriksson   Tue, 06 Nov 2018 20:53:01 +0100

What is the output of the following command:
 apt-listchanges --profile=apt --dump-seen | grep -E '(util-linux|lvm2)'


Regards,
robert

Bug#908551: apt-listchanges: apt update hangs if no mail system

[Robert Luberda]

Michael Meskes pisze:
> On Tue, Sep 11, 2018 at 11:11:34PM +0200, Robert Luberda wrote:
>> reassign 908551 citadel-server  902-4
>> ...
>> like this (BTW. I've just temporaily installed the latest version of
>> citadel-server, trying to reproduce the bug, but its sendmail command
>> just fails, not hangs):
> 
> Sorry, I don't understand this. Why do you reassign to citadel-server when 
> your
> test shows citadel-server works correctly?

I reassigned it, because the bug reporter had citadel-server installed,
and it appeared to hang for him for some reason.

(Also I wouldn't agree it worked correctly, as it showed some failure
message instead of sending my e-mail. Most probably it was a matter of
some missing configuration, but I didn't have time to spend on it; I
only installed the package, checked that I couldn't reproduce the exact
issue from the bug report, and then uninstalled it.)

Regards,
robert

Bug#910590: bsd-mailx: error about missing command "jalias"

[Robert Luberda]

Francesco Potortì writes:

> $ echo test | mailx pot
> Unknown command: "jalias"
> $
> 
> The mail is sent, ma an error is signaled

Please check your ~/.mailrc and /etc/mail.rc files. I guess one of them
contains the "jalias" command... If you remove it or replace with
"alias", then the error should be gone.

Regards,
robert

Bug#908551: apt-listchanges: apt update hangs if no mail system

[Robert Luberda]

RJ pisze:

An issue in apt-listchanges is that handling of Ctrl-C does not work,
when /bin/sh points to dash; this is visible here:

> ^CE: Sub-process /usr/bin/apt-listchanges --apt || test $? -lt 10
> received signal 2.
> E: Failure running script /usr/bin/apt-listchanges --apt || test $? -lt 10

Pressing Ctrl-C causes apt-listchanges to exit with code 10, so the
above command should return 0, but... dash as parent process receives
SIGINT as well, what ends up with returning code 130.
bash works differently and returns 0, but I guess unfortunately dash
behavior's is correct according to POSIX. I will try to fix this inside
apt-listchanges somehow.

Regards,
robert

Bug#908551: apt-listchanges: apt update hangs if no mail system

[Robert Luberda]

RJ writes:

reassign 908551 citadel-server  902-4
thanks

> 
> Versions of packages apt-listchanges suggests:
> ii  citadel-server [mail-transport-agent]  902-4


> -- Desription of problem:
> When performing a apt update today (Monday, September 10, 2018) there
> was a news bit about the Apache/2 package that required reading and
> confirmation.
> Upon confirmation the apt-listchanges script returned to attempt to send
> mail to the configured user email (which is just 'root' in this case.)

Yes, apt-listchanges tries to execute `/usr/sbin/sendmail -oi -t'
command. It skips sending mails if the command does not exist.

> IF a system either does not have a mail service installed 

You have citadel-server installed, that provides mail service, so it's
not the case.

> or is
> impropery configured, apt-listchanges will hang while attempting to send
> an email before it allows the update process to continue.

IMHO if a mail service is improperly configured than it should fail
instead of hanging... Then apt-listchanges would display warning message
like this (BTW. I've just temporaily installed the latest version of
citadel-server, trying to reproduce the bug, but its sendmail command
just fails, not hangs):

  apt-listchanges: Mail to root: apt-listchanges: changelogs for vox
citmail: can't connect: No such file or directory
apt-listchanges warning: Cannot send mail to root: Command
'['/usr/sbin/sendmail', '-oi', '-t']' returned non-zero exit status 3.

Then apt-listchanges would continue. I agree, that apt-listchanges could
have set some timeout for  the sendmail command, and I can try to add
such timeout in next version.

> 
> ** I was able to work around this by editing the apt-listchanges script
> file by commenting out the offending Mail section as follows:

A simpler solution would be to run `dpkg-reconfigure apt-listchanges'
and give an empty list of e-mail addresses in the question related to
mail addresses.

> As with the other related 'Wishlist' bug reports for apt-listchanges,
> there should be additional prompts added to this file to:
> 
> (1) Ask if the user wants to mail a report (default N). 

This is done during apt-listchanges installation time, providing that
your debconf settings allows the question about e-mail address list to
appear.

> - Also notify
> user that the process may hang if no mail subsystem exists (which I
> believe is now the case starting with stretch) or is improperly
> configured. (Just 'root')

I think that the sendmail command should hang when misconfigured or for
any other reason - that's why I am reassigning the bug report to
citadel-server. (BTW. I am not sure if this is possible in case of
citadel-server, but maybe sendmail command was trying to send e-mails
via network? How long did you wait before deciding that the command hung?).

> 
> (2) See other 3 reports related to MAIL for additional prompt suggestions.

Could you please be more specific? Which other 3 bug reports are related
to this one?

Regards,
robert

Bug#907620: lintian: Bogus 'unknown-runtime-tests-feature' tags

[Robert Luberda]

Package: lintian
Version: 2.5.99
Severity: minor

According to 
https://salsa.debian.org/ci-team/autopkgtest/blob/master/doc/README.package-tests.rst,
 
there is one defined feature, namely:

  test-name
Set an explicit test name for the log heading and the summary file for a 
Test-Command: 
inline test. When not given, these are enumerated like command1

However lintian does not seem to know this, as it produces the following
output while checking upx-ucl 3.95-1:

P: upx-ucl source: unknown-runtime-tests-feature test-name=bug775451-check1 
paragraph starting at line 3
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug775451-check2 
paragraph starting at line 6
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug775455-check 
paragraph starting at line 9
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug873260-check 
paragraph starting at line 12
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug882401-check 
paragraph starting at line 15
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug899190-check1 
paragraph starting at line 18
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug899190-check2 
paragraph starting at line 21
P: upx-ucl source: unknown-runtime-tests-feature test-name=bug907426-check 
paragraph starting at line 24


Regards,
robert

-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Init: systemd (via /run/systemd/system)

Versions of packages lintian depends on:
ii  binutils   2.31.1-4
ii  bzip2  1.0.6-9
ii  diffstat   1.61-1+b1
ii  dpkg   1.19.0.5+b1
ii  file   1:5.34-2
ii  gettext0.19.8.1-7
ii  intltool-debian0.35.0+20060710.4
ii  libapt-pkg-perl0.1.34
ii  libarchive-zip-perl1.63-1
ii  libclass-accessor-perl 0.51-1
ii  libclone-perl  0.39-1
ii  libdpkg-perl   1.19.0.5
ii  libemail-valid-perl1.202-1
ii  libfile-basedir-perl   0.08-1
ii  libipc-run-perl20180523.0-1
ii  liblist-moreutils-perl 0.416-1+b3
ii  libparse-debianchangelog-perl  1.2.0-12
ii  libtext-levenshtein-perl   0.13-1
ii  libtimedate-perl   2.3000-2
ii  liburi-perl1.74-1
ii  libxml-simple-perl 2.25-1
ii  libyaml-libyaml-perl   0.72+repack-1
ii  man-db 2.8.4-2
ii  patchutils 0.3.4-2
ii  perl [libdigest-sha-perl]  5.26.2-7
ii  t1utils1.41-2
ii  xz-utils   5.2.2-1.3

Versions of packages lintian recommends:
pn  libperlio-gzip-perl  

Versions of packages lintian suggests:
pn  binutils-multiarch 
ii  dpkg-dev   1.19.0.5
ii  libhtml-parser-perl3.72-3+b2
ii  libtext-template-perl  1.53-1

-- no debconf information

Bug#847786: solid-pop3d: Does not work with systemd. Is inetd mandatory?

[Robert Luberda]

Ludovic Rousseau writes:

Hi,
> 
> I just installed solid-pop3d in my Debian stable (jessie 8.6) system.
> 
> It looks like the pop3 server is NOT started when systemd is used.
> I use the default package configuration so "inetd". I have not tried
> with the "daemon" configuration.

The "deamon" configuration should suit you. It causes solid-pop3d to run
as a standalone daemon, without using inetd. It should work with systemd
as well, so I am closing this bug report.
> 
> It is possible to use this package on a systemd system (default init
> system on Debian now)?
> If yes can you provide the systemd configuration files?

The files are autogenerated, based on /etc/init.d/solid-pop3d.

Regards,
robert

Bug#677944: bas-mailx priority has changed

[Robert Luberda]

Daniel Pocock writes:

Hi,

> 
> I've reopened this bug because of two things:
> 
> a) the priority of bsd-mailx was lowered[1] to optional in 2016 (Robert,
> can you comment on that?)

The priority was actually lowered by ftpmasters in the override file
(see [1]), most probably because bsd-mailx "pulls in an MTA which
is also no longer included by default [in standard installation]" [2].
The change in bsd-mailx in fact changed nothing, as it was only to
synchronize the value of the field in the package with the override.

[1] https://wiki.debian.org/FtpMaster/Override
[2] https://lists.debian.org/debian-user/2017/07/msg00459.html

> 
> b) consequently, mailutils is now being installed by default in stretch
> and it doesn't respect /etc/mailname - #253513[2]

exim4-base recommends pure virtual mailx package, and apt somehow
chooses mailutils to satisfy that recommendation. I think this might be
a kind of a random choice, because IMHO nothing in the dependencies
system seems to tell apt which package should it choose. Honestly I
cannot see any way to fix this choice at the bsd-mailx package side.

Regards,
robert

Bug#902253: enigmail: Incorrect Polish translation for "Sign message"

[Robert Luberda]

Package: enigmail
Version: 2:2.0.7-6
Severity: normal
Tags: l10n

For some reason Polish translation of "Sign message" menu position 
has been (recently?) changed from "Podpisz wiadomość" into "Znak Wiadomość" 
which I don't think any Pole can understand, unless they can restore original
English from this  pseudo-Polish after realizing that English noun "sign" 
can be translated into "znak" (e.g. "a road sign" is "znak drogowy").

Regards,
robert


-- System Information:
Debian Release: buster/sid
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages enigmail depends on:
ii  gnupg2.2.8-3
ii  gnupg-agent  2.2.8-3
ii  gnupg2   2.2.8-3
ii  gpg-agent [gnupg-agent]  2.2.8-3
ii  icedove  1:52.8.0-1~deb9u1
ii  thunderbird  1:52.8.0-1

Versions of packages enigmail recommends:
ii  pinentry-gnome3 [pinentry-x11]  1.1.0-1+b1

enigmail suggests no packages.

-- no debconf information

Bug#898749: myspell-pl: Convert into dummy transitional package depending on hunspell-pl

[Robert Luberda]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

reassign 898749 src:libreoffice-dictionaries
retitle 898749 Please provide dummy transitional myspell-pl package
thanks

Pander pisze:
> Package: myspell-pl Version: 20161007-1 Severity: normal Tags:
> l10n
> 
> Convert myspell-pl into dummy transitional package depending on 
> hunspell-pl and gradually phase out myspell-pl.

Actually hunspell-pl conflicts with myspell-pl. As the conflict needs
to be dropped anyway (or actually replaced with versioned Breaks), I
think that the libreoffice-dictionaries can provide the transational
package for myspell-pl instead. Once this is done, I can drop
myspell-pl from ipolish source package.

Regards,
robert



-BEGIN PGP SIGNATURE-
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=1C5m
-END PGP SIGNATURE-

Bug#901391: wpolish: Convert dictionaries-common file to UTF-8

[Robert Luberda]

Pander pisze:

Hi,

> Why is this package depending on the perl package? I see no reason.


The dependency is added by dh_perl due to debconf's config script put
into binary package by dictionary-common-dev.

Regards,
robert

Bug#888297: closed by Robert Luberda <rob...@debian.org> (Bug#888297: fixed in p7zip 16.02+dfsg-5)

[Robert Luberda]

Antoine Beaupré writes:

Hi,

>> The check for cur against kNumItems is missing, not sure this can
>> cause any further problem.
> 
> I concur: the original researcher explicitly sent me a patch that checks
> the `cur` counter as well.

Thanks, I'm just uploading -6 with updated patch.

Regards,
robert

Bug#886462: wajig: "wajig install" actually removes packages

[Robert Luberda]

Package: wajig
Version: 2.18.1
Severity: important

Hi, 

"wajig install package" tries to remove a bunch of unrelated packages,
for example:

env LC_ALL=C wajig install apt-src
Reading package lists... Done
Building dependency tree   
Reading state information... Done
The following packages will be REMOVED:
  gir1.2-wnck-3.0 libbrlapi0.6 libdotconf0 liblouis-data liblouis14 libspeechd2 
libwnck-3-0 libwnck-3-common
  libxres1 orca python3-brlapi python3-louis python3-speechd speech-dispatcher 
speech-dispatcher-audio-plugins
The following NEW packages will be installed:
  apt-src
0 upgraded, 1 newly installed, 15 to remove and 10 not upgraded.


I've seen this behaviour for ages, but I cannot get used to it, because
in my opinion this is just a classic WTF.
Every time I try use "wajig install" I'm ending up with "sudo apt-get install" 
because it's much simpler  than checking why "apt-get install --auto-remove" 
(that is called by wajig) thinks that it is OK to remove my packages. apt-get
 *made* wrong  choices in the past, that's why I prefer to remove packages 
manually  via e.g. synaptic or orphaner (what also means that I never use
"wajig remove" or "wajig purge", which - judging from the source code -
share the same counterintuitive behaviour with "wajig install").


Regards,
robert

Bug#881536: ffmpeg: Breaks sound in kodi

[Robert Luberda]

Viktor Malyarchuk pisze:

Hi

> just want to confirm that
> 
> gcc -shared shim.c -olibshim.so
> LD_PRELOAD=$(pwd)/libshim.so kodi-standalone
> 
> indeed fixed the audio problem with kodi.

I can confirm that too.

Thanks a lot,
robert

Bug#881536: ffmpeg: Breaks sound in kodi

[Robert Luberda]

Package: ffmpeg
Version: 7:3.4-2
Severity: important

The latest ffmpeg makes sound in kodi to be scratchy, i.e. containing some 
additional noise that makes watching videos in kodi uncomfortable/annoying.

Downgrading ffmpeg and dependent libraries to version 3.3.4-2+b3 fixes the
issue (makes sound in kodi clear).

Regards,
robert


-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'stable-updates'), (990, 
'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.13.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages ffmpeg depends on:
ii  libavcodec577:3.4-2
ii  libavdevice57   7:3.4-2
ii  libavfilter67:3.4-2
ii  libavformat57   7:3.4-2
ii  libavresample3  7:3.4-2
ii  libavutil55 7:3.4-2
ii  libc6   2.24-17
ii  libpostproc54   7:3.4-2
ii  libsdl2-2.0-0   2.0.7+dfsg1-3
ii  libswresample2  7:3.4-2
ii  libswscale4 7:3.4-2

ffmpeg recommends no packages.

Versions of packages ffmpeg suggests:
pn  ffmpeg-doc  

-- no debconf information

Bug#872926: sysstat: running sadc results in segmentation fault

[Robert Luberda]

Jamie McClelland writes:

> I started getting email messages from the sysstat cron job reporting the
> segmentation fault.

When did you start receiving those e-mails? Just after upgrade systat to
11.5.7 or later i.e. did version 11.5.7 work, but then stopped for some
reason?

> 
> root@turkey:~# /usr/lib/sysstat/sadc -F -L -S DISK 1 1 /var/log/sysstat
> Segmentation fault
> root@turkey:~#
> 
> Here's an strace:
> 
> open("/etc/mtab", O_RDONLY) = 3
> fstat(3, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
> read(3, "sysfs /sys sysfs rw,nosuid,nodev"..., 1024) = 1024
> [...]
> statfs("0", 0x7ffef1ab0890) = -1 ENOENT (No such file or 
> directory)
> read(3, "y2/l/7EEOV2FTKYKCYQJPMOALKJSL3M:"..., 1024) = 1024
> --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0x1} ---

Looks like the last file read was /etc/mtab (which should be a symlink
to /proc/self/mtab). Even though the beginning of the file looks pretty
usual, then something in this file causes sadc to crash.

Could you please send the contents of the /etc/mtab file, especially the
lines around the "y2/l/7EEOV2FTKYKCYQJPMOALKJSL3M" string? In case it
contain data you don't want to share into public, e.g. server names or
paths on remote servers, please obfuscate it.

Thanks a lot,
robert

Bug#871647: upx-ucl: uses convenience copy of LZMA library

[Robert Luberda]

Thorsten Glaser writes:

> 
> Apparently upx-ucl reverted from using the system LZMA library
> to its own.

First of all, lzma-sdk is not designed to be used as a library, so in
fact there is no such thing like "system LZMA library". Moreover upx is
not the only package that includes its own copy of LZMA, see e.g.
https://codesearch.debian.net/search?q=__LZMA_ENCODER_H=1

> This, if not justified, is a Policy violation, for
> various reasons, including maintainability and security.
> 
> Neither the changelog entry nor README.{Debian,source} contain
> any justification, however.

The justification is simple and obvious: current upx does not compile
with the current lzma-dev.

upx requires the following files:

  compress_lzma.cpp:#include "C/Common/MyInitGuid.h"
  compress_lzma.cpp:#include "C/7zip/Compress/LZMA/LZMAEncoder.h"
  compress_lzma.cpp:#include "C/Common/Alloc.cpp"
  compress_lzma.cpp:#include "C/Common/CRC.cpp"
  compress_lzma.cpp:#include "C/7zip/Common/OutBuffer.cpp"
  compress_lzma.cpp:#include "C/7zip/Common/StreamUtils.cpp"
  compress_lzma.cpp:#include "C/7zip/Compress/LZ/LZInWindow.cpp"
  compress_lzma.cpp:#include "C/7zip/Compress/LZMA/LZMAEncoder.cpp"
  compress_lzma.cpp:#include "C/7zip/Compress/RangeCoder/RangeCoderBit.cpp"
  compress_lzma.cpp:#include "C/7zip/Compress/LZMA_C/LzmaDecode.h"
  compress_lzma.cpp:#include "C/7zip/Compress/LZMA_C/LzmaDecode.c"
  stub/src/c/lzma_d_c.c:#include "C/7zip/Compress/LZMA_C/LzmaDecode.h"
  stub/src/c/lzma_d_c.c:#include "C/7zip/Compress/LZMA_C/LzmaDecode.c"

while lzma-dev in Debian provides only:

  /usr/include/lzma/LzmaDec.c
  /usr/include/lzma/LzHash.h
  /usr/include/lzma/LzFind.h
  /usr/include/lzma/LzmaDec.h
  /usr/include/lzma/LzmaEnc.h
  /usr/include/lzma/7zVersion.h
  /usr/include/lzma/LzmaEnc.c
  /usr/include/lzma/LzFind.c
  /usr/include/lzma/Types.h

Yes, I know that I can try to request adding new files into lzma-dev, as
it was done in past, see https://bugs.debian.org/452817, but at that
time upx source did not include copy of lzma, and now it does.

Also let me quote the following upstream note from
stub/src/c/Makevars.lzma file in upx source code:

  # UPX unconditionally uses its own version in src/lzma-sdk because
  # that version works fine since 2006 and that is the only version
  # that is actually sufficiently tested!!!

Regards,
robert

Bug#869938: Raising severity to serious since g-s-d 3.24 has been uploaded to unstable

[Robert Luberda]

severity 869938 important
thanks

> I have the package ready, I will upload it shortly.

I'm lowering the severity so that the package won't migrate to testing
too early.

Regards,
robert

Bug#869938: Raising severity to serious since g-s-d 3.24 has been uploaded to unstable

[Robert Luberda]

Jeremy Bicha pisze:
> Control: severity -1 serious
> 
> gnome-settings-daemon 3.24 has been uploaded to unstable. We intend to
> NMU your package today so that the transition is not delayed.

I have the package ready, I will upload it shortly.


Regards,
robert

Bug#781987: Regarding bug #781987 - dwww: All links are broken after clean install

[Robert Luberda]

David R Forrest writes:

> Arjan's one-line patch to /usr/sbin/dwww-convert corrects this for me.
>
> I get the expected results for /usr/share/doc files, and no longer get the 
> 500 error document in the browser nor the "AH02429: Response header name 
> 'Last modified' contains invalid characters, aborting request" message in 
> error.log

This patch was applied to dwww a few months ago, and the current stable
release of Debian (stretch) already contains a fixed version - please
consider upgrading to it. (However this particular bug is still open,
because it originally referred to quite a different issue).

For jessie it most probably won't be fixed at all, as the oldstable
release managers have not been able to approve such a small change since
May, see #862363.

Regards,
robert

Bug#868695: systemd: leaves empty LC_CTYPE what breaks X11 ssh password prompt

[Robert Luberda]

Michael Biebl writes:

Hi,

Sorry for not replying for your previous e-mails, I was too tired
yesterday after work, and planned to do this today, but probably it
doesn't make sense anymore as you've found the commit that caused the
issue. Anyway I was pretty sure that the LC_CTYPE settings are
relevant, because before reporting the bug I tried to comment them
out, what fixed the problem. But than I noticed that those settings
were present also in the previous version, but unfortunately missed
the difference in the Exec line of getty@.service, so I thought that
some change in systemd internals prevented it from clearing the
environment :(

> Am 17.07.2017 um 20:54 schrieb Robert Luberda:
>> Yes, LC_CTYPE and other LC_ variables are empty, because they are
>> set like this in /lib/systemd/system/getty@.service.
> 
> Afaics, this was broken by 
> https://github.com/systemd/systemd/pull/6023
> 
> Robert, can you try reverting the commit 
> https://github.com/systemd/systemd/commit/db6aedab9292678918f15807a0d835be35511667
>
>  and report back.

I've just replaced the getty@.service file with the version coming
from systemd 233-10, and yes, I confirm that doing this followed by
rebooting the system fixed the bug: LC_CTYPE is unset, and ssh client
shows the X11 gui prompt.

> I've created an upstream bug report in the mean time at 
> https://github.com/systemd/systemd/issues/6407
> 

Oh, I've just checked that according to
https://codesearch.debian.net/search?q=OPTION+lc-ctype
the empty LC_CTYPE might break a few other packages: kwalletcli,
scoute, and nss-passwords as well.


Thanks a lot,
robert

Bug#868695: systemd: leaves empty LC_CTYPE what breaks X11 ssh password prompt

[Robert Luberda]

Package: systemd
Version: 234-1
Severity: critical
Justification: breaks unrelated software

I usually login on tty1 via getty and then switch to X via startx.
After recent upgrade of systemd  I can no longer ssh to any 
host in  X11, because gpg-agent (which is used instead of ssh-agent) 
fails to prompt for password, and the following error is shown instead:

 sign_and_send_pubkey: signing failed: agent refused operation
 Permission denied (publickey).

strace -f on gpg-agent process showed me this:

[pid  1794] <... read resumed> "OPTION lc-ctype=", 1002) = 16
[pid  1792] write(12, "\n", 1 
[pid  1794] read(6,  
[pid  1792] <... write resumed> )   = 1
[pid  1794] <... read resumed> "\n", 986) = 1
[pid  1792] read(9,  
[pid  1794] write(10, "ERR 536871188 IPC syntax error <"..., 81) = 81
[pid  1792] <... read resumed> "ERR 536871188 IPC syntax error <"..., 1002) = 81


Yes, LC_CTYPE and other LC_ variables are empty, because they are set
like this in /lib/systemd/system/getty@.service.
Downgrading systemd to 233-10 solves the bug for me (even though
getty@.service still contains the empty LC_ variables, but fortunatelly they 
are not propagated to my terminals).

Regards,
robert

-- Package-specific info:

-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.11.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE= (charmap=UTF-8), LANGUAGE= (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages systemd depends on:
ii  adduser 3.115
ii  libacl1 2.2.52-3+b1
ii  libapparmor12.11.0-6
ii  libaudit1   1:2.7.7-1+b1
ii  libblkid1   2.29.2-2
ii  libc6   2.24-12
ii  libcap2 1:2.25-1
ii  libcryptsetup4  2:1.7.3-4
ii  libgcrypt20 1.7.8-2
ii  libgpg-error0   1.27-3
ii  libidn111.33-1
ii  libip4tc0   1.6.1-2
ii  libkmod224-1
ii  liblz4-10.0~r131-2+b1
ii  liblzma55.2.2-1.3
ii  libmount1   2.29.2-2
ii  libpam0g1.1.8-3.6
ii  libseccomp2 2.3.1-2.1
ii  libselinux1 2.6-3+b2
ii  libsystemd0 234-1
ii  mount   2.29.2-2
ii  procps  2:3.3.12-3
ii  util-linux  2.29.2-2

Versions of packages systemd recommends:
ii  dbus1.10.20-1
ii  libpam-systemd  234-1

Versions of packages systemd suggests:
ii  policykit-10.105-18
pn  systemd-container  

Versions of packages systemd is related to:
pn  dracut   
ii  initramfs-tools  0.130
ii  udev 234-1

-- no debconf information

Bug#866583: apt-listchanges phones home

[Robert Luberda]

David Weinehall writes:
> 
> Introducing a "feature" like this should only be done if the meta-data
> is mirrored together with the packages.

In version 3.14 I added `--no-network' option to disable this behavior.

Regards,
robert

Bug#867743: devscripts: Upgrade fails due to conflict with debdelta

[Robert Luberda]

Package: devscripts
Version: 2.17.7
Severity: serious

Upgrade fails with:

Preparing to unpack .../devscripts_2.17.7_amd64.deb ...
Unpacking devscripts (2.17.7) over (2.17.6) ...
dpkg: error processing archive 
/var/cache/apt/archives/devscripts_2.17.7_amd64.deb (--unpack):
 trying to overwrite '/usr/share/man/man1/debpatch.1.gz', which is also in 
package debdelta 0.59
dpkg-deb: error: subprocess paste was killed by signal (Broken pipe)
Errors were encountered while processing:
 /var/cache/apt/archives/devscripts_2.17.7_amd64.deb

Regards,
robert

Bug#867511: ucf: Sometimes skips file dialog for files generated during dpkg-reconfigure

[Robert Luberda]

Package: ucf
Version: 3.0036
Severity: important

Hi,

apt-listchanges uses debconf-based interface to generate a new version
of config file, and then calls ucf to update the config file with the
new version. I've just noticed that this does not work properly when
the generated file is the same as the file that was previously chosen by
a user to be kept unchanged.

To explain this behaviour in details let me use the following example 
steps the user can made:

1. Let's assume that the user changes something in listchanges.conf
   file manually, e.g.
  echo "someoption = value" >> /etc/apt/listchanges.conf

2. Then he or she  uses `dpkg-reconfigure apt-listchanges' command to
   change some apt-listchanges settings. For simplicity let's assume that the
   dpkg-reconfigure does something like:
 sed -e 's/value/VALUE/' < /etc/apt/listchanges.conf > 
/etc/apt/listchanges.conf.new

3. Now dpkg-reconfigure calls ucf:
 /usr/bin/ucf --debconf-ok /etc/apt/listchanges.conf.new 
/etc/apt/listchanges.conf
   Ucf shows a file modification prompt, and that's OK.

4. But now let's assume that the the user selects "keep the local version 
currently
   installed" for example by mistake, because this is the default option.

5. If he or she wants to correct the mistake, runs the dpkg-reonfigure
   command again, and selects exactly the same options, what makes 
   dpkg-reconfigure to create exactly the same new config file as previously. 
   Obviously the same ucf command will be called, but this time 
   *ucf will happily exit without prompting the user* for anything - and this 
   is the bug this report is about.

Regards,
robert


-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (990, 'unstable-debug'), (990, 'unstable'), (990, 'testing'), 
(990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.11.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages ucf depends on:
ii  coreutils  8.26-3
ii  debconf1.5.62

ucf recommends no packages.

ucf suggests no packages.

-- debconf information:
  ucf/title:
* ucf/show_diff:
* ucf/changeprompt_threeway: keep_current
  ucf/conflicts_found:
  ucf/changeprompt: keep_current

Bug#862363: jessie-pu: package dwww/1.12.1+deb8u1

[Robert Luberda]

reopen 862363 !
thanks

Cyril Brulebois writes:

>> Could you please approve this change and allow Adrian to proceed with
>> the NMU?
>>
>> It fixes a pretty old bug in dwww that was recently made visible (and
>> thus made dwww mostly unusable) due to the security upload of apache2
>> into jessie.
> 
> This is the kind of things that should have been in the pu request. This
> should also be mentioned in the changelog. Fixing bugs is great, but
> providing explanations while doing so is even better.
> 

I can see that Adrian has closed the bug report, so I've just prepared
another version of the patch:


diff --git a/debian/changelog b/debian/changelog
index d30cd60..48f6a8d 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,11 @@
+dwww (1.12.1+deb8u1) jessie; urgency=medium
+
+  * Fix an old typo in the `Last-Modified' header name that prevents dwww
+from working correctly on systems running the latest available jessie
+version of apache2, which as a part its security update for
CVE-2016-8743
+started enforcing HTTP headers conformance with the appropriate
standards
+(closes: #850016, #850885).
+
 dwww (1.12.1) unstable; urgency=medium

   * apache.conf: add an `Alias /dwww /var/www/dwww' line to accommodate
diff --git a/scripts/dwww-convert b/scripts/dwww-convert
index a9792c5..d296d32 100755
--- a/scripts/dwww-convert
+++ b/scripts/dwww-convert
@@ -327,7 +327,7 @@ sub PrintHeaders() { # {{{
 print "Content-type: $mime_type" . (defined $mime_charset ? ";
charset=$mime_charset\n" : "\n");
 my @stat = stat( $filename );
 my $mtime = $stat[9];
-print "Last modified: " . gmtime($mtime) . "\n";
+print "Last-modified: " . gmtime($mtime) . "\n";
 print "Content-Disposition: inline; filename=\"$base_name\"\n";
 print "\n";
 } # }}}


Would this be accepted? If yes, could you please let me know how should
I upload this, as I haven't uploaded anything to stable for ages?

Regards,
robert

Bug#867214: git-buildpackage: Strange format of commit message added by import-orig

[Robert Luberda]

Guido Günther writes:

> Using past tense was indeed a bad choice.  I deliberately avoided the
> terme "merge" here since we're not using the git merge machinery but
> build the tree ourself (see --merge-mode in gbp-import-orig(1)). 

I used "merge" in my proposal, because technically the resulting commit
is a merge commit, as it contains two parents, and the way it was
produced could still be deducted from the second part of the commit message.
However I won't insist on that, i.e. using "update" instead of "merge"
is OK, as long as the commit message does not start with a sentence
artificially split in the middle with an empty line.

> What about:>
>   Update upstream source from tag 'upstream/20170702'
>  
>   Update to upstream version 20170702
>   with Debian dir 43f841d7950f0587745d57c8c87fd79e1aed60dd.
> 

It looks great to me!

Thanks,
robert

Bug#867214: git-buildpackage: Strange format of commit message added by import-orig

[Robert Luberda]

Package: git-buildpackage
Version: 0.8.17
Severity: normal

Hi,

gbp import-orig stared producing merge commit messages like 
this one:


Updated version 20170702 from 'upstream/20170702'

with Debian dir 43f841d7950f0587745d57c8c87fd79e1aed60dd


which I don't really like for the three following reasons:

1. The commit message uses a really bizzare format of:
   
A description of the change split for some

reason in the middle of a sentence

instead of:
   
Short summary of the change

More detailed description possibly consisting of several
lines or even paragraphs with each sentence ended with 
a trailing dot.

2. I am not native speaker, and maybe that's why I have an impression
that a small word "to" is missing in "Updated version 20170702". But to
be honest I can't really see the need of duplicating the version number
in the same line. Prevously the version was mentioned once in the summary, 
and once in the long description, what IMHO looked much better:

Merge tag 'upstream/20161007'

Upstream version 20161007

3. The commit message uses past tense ("Updated"), where previously infinitive
was used (e.g. "Merge").



It would be nice if all those three items could be fixed. In my opinion
something like this:

Merge tag 'upstream/20170702'

Update to upstream version 20170702
with Debian dir 43f841d7950f0587745d57c8c87fd79e1aed60dd.

would but better, but as I wrote I am not a native.

Regards,
robert
  

Bug#866728: lintian: spell checks for scripts or alternatively POT files?

[Robert Luberda]

Package: lintian
Version: 2.5.51
Severity: wishlist

Hi

Yesterday I received a bug #866644 about apt-listchanges printing "retrive" 
instead of "retrieve" in one of messages. If apt-listchanges was written 
in C, lintian would surely have complained about spelling typo in binary.
However apt-listchanges is just a Python script with modules...

It would be nice if lintian could somehow check scripts for spelling typos, 
just like it does for compiled binaries. I know that it might be hard to do
correctly. My first impression was that it could check all words inside 
the "" or '' quotes, ignoring any words starting with $ or %, but 
than I recalled things like eval(), os.system() in Python or similar
constructs in other script languages. 
But maybe it would be possible to spell check strings marked for translation 
between `_(', and `)'? Or alternatively check msgids in the POT files? What
do you think?

Regards,
robert

-- System Information:
Debian Release: buster/sid
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Bug#866583: apt-listchanges phones home

[Robert Luberda]

Jakub Wilk writes:
> Package: apt-listchanges
> Version: 3.12
> 
> For some packages, apt-listchanges calls "apt-get -qq changelog
> $package=$version", 


Yes, I've recently merged a patch from Ubuntu, because I think this is a
pretty nice feature to be able to see changelogs even if a package
itself does not have one, e.g. because its documentation directory is a
symlink to directory provided by other dependent package. Another reason
is to allow newer versions of apt-listchanges to be automatically pulled
by Ubuntu to make bug reports in their Launchpad more valuable to me.

However after merging the patch, I've found that it calls `apt-get
changelog' a way too often. I've already have fix for this, which I'm
currently testing and plan to upload during the weekend. With the fix
the apt-get command will *almost* never be called during the normal
upgrades (but e.g. will be called if somebody runs e.g. `apt-listchanges
-a single-package-without-changelog.deb').

> which downloads changelog from
> http://metadata.ftp-master.debian.org/.
> I did not agree to disclosing
> which packages I install to this party. 

Do you mean to Debian? Is it less trustworthy than the mirror you
download .deb files during upgrades from? :)

> Please disable this misfeature.

I can add config file option for this.

Regards,
robert

Bug#866358: src:apt: Typo in urgency in changelog entry

[Robert Luberda]

Source: apt
Version: 0.9.7.8
Severity: minor

While working on apt-listchanges I noticed that apt's changelog file contains 
an extra `i' in the following line:
 apt (0.9.7.8) unstable; urgency=criticial
   ^

This typo exposes a bug in apt-listchanges that prevents command like
`apt-listchanges --show-all apt_*.deb' from showing the changelog. 
The bug will be fixed in next upload of apt-listchanges, but I think
it would be nice have the typo fixed as well.

Regards,
robert


-- System Information:
Debian Release: buster/sid
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Bug#865748: popularity-contest: Should handle dpkg diversions

[Robert Luberda]

Package: popularity-contest
Version: 1.65
Severity: wishlist
Tags: patch

It would be nice if popularity-contest could handle diversions.
Otherwise the same file can be reported for different packages, e.g.:

  $ perldoc >/dev/null 2>&1
  $ /usr/sbin/popularity-contest | grep perldoc
  1498305600 1498089600 perl-doc /usr/bin/perldoc
  1498305600 1498089600 perl /usr/bin/perldoc

The following patch handles this by choosing the simplest approach,
namely by ignoring diverted files unless they were diverted by the
package being worked on.

--- popularity-contest.old
+++ popularity-contest
@@ -107,6 +107,16 @@
   }
 }
 
+# List files diverted by dpkg
+my %diverted;
+open DIVERSIONS, "LC_ALL=C dpkg-divert --list|";
+while ()
+{
+  next unless /^(?:local )?diversion of (\S+) to (?:\S+)(?: by (\S+))?\s*$/;
+  $diverted{$1} = defined $2 ? $2 : ""
+}
+close DIVERSIONS;
+
 # Read dpkg database of installed packages
 open PACKAGES, "dpkg-query --show --showformat='\${status} \${package}\\n'|";
 while ()
@@ -127,6 +137,7 @@
 && ! m{/lib/.+-.+-.+/}
 || defined $mapped{$_} )
   && -f $_);
+  next if defined $diverted{$_} and $diverted{$_} ne $pkg;
   my($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
   $atime,$mtime,$ctime,$blksize,$blocks)
 = stat;

Regards,
robert

-- System Information:
Debian Release: 9.0
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages popularity-contest depends on:
ii  debconf [debconf-2.0]  1.5.61
ii  dpkg   1.18.24

Versions of packages popularity-contest recommends:
ii  cron [cron-daemon]  3.0pl1-128+b1
ii  gnupg   2.1.18-8
ii  postfix [mail-transport-agent]  3.2.2-1

Versions of packages popularity-contest suggests:
ii  anacron  2.3-24

-- debconf information excluded

Bug#865730: popularity-contest: Checking /proc/*/maps seems to be ineffective

[Robert Luberda]

Package: popularity-contest
Version: 1.65
Severity: normal

popularity-contest has some code to scan /proc/*/maps files to check
for currently used programs. However popularity-contest itself is
usually run from the cron job as the nobody user, who does not have
permission to read those /proc/*/maps files...

I think that either the support for reading the maps files should be
dropped or popularity-contest should be run as root (BTW. it would make
sense to run popcon-upload as nobody instead of root), however the
second option would change the contents of the reports, because 
currently for some reason  the multi-arch libs are ignored via the 
"! m{/lib/.+-.+-.+/}" (which BTW. does not work on hurd-i386),
except for files coming from /proc/*/maps. This causes most library
packages to be reported as `no-files' except for libperl5.24 used
by the popularity-contest process itself, e.g. my recent log file
contains:

  149796 1496534400 libperl5.24 
/usr/lib/x86_64-linux-gnu/perl/5.24.1/auto/File/Glob/Glob.so

Running popularity-contest as root would cause more *.so files to be
reported because of the /proc/*/maps, for example libcap2 would be
sent as:
  1498262400 1490616000 libcap2 /lib/x86_64-linux-gnu/libcap.so.2.25
instead of :
  0 0 libcap2 
(Most probably similar result would be possible to gain even when
running as nobody if the exclusion of files matching to multi-lib pattern 
is removed).

Regards,
robert

-- System Information:
Debian Release: 9.0
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages popularity-contest depends on:
ii  debconf [debconf-2.0]  1.5.61
ii  dpkg   1.18.24

Versions of packages popularity-contest recommends:
ii  cron [cron-daemon]  3.0pl1-128+b1
ii  gnupg   2.1.18-8
ii  postfix [mail-transport-agent]  3.2.2-1

Versions of packages popularity-contest suggests:
ii  anacron  2.3-24

-- debconf information excluded

Bug#865728: popularity-contest: Should ignore /usr/lib/mime/packages

[Robert Luberda]

Package: popularity-contest
Version: 1.65
Severity: minor

I've noticed that my /var/log/populatiry-contest contains entries like:

  149796 1490616000 mime-support /usr/lib/mime/packages/mime-support
  149796 1490616000 less /usr/lib/mime/packages/less
  149796 1490616000 tar /usr/lib/mime/packages/tar

The files are accessed most probably by update-mime(8) only via trigger
during packages installation, so including them in popularity-contest's
report does not make much sense.

Regards,
robert

-- System Information:
Debian Release: 9.0
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages popularity-contest depends on:
ii  debconf [debconf-2.0]  1.5.61
ii  dpkg   1.18.24

Versions of packages popularity-contest recommends:
ii  cron [cron-daemon]  3.0pl1-128+b1
ii  gnupg   2.1.18-8
ii  postfix [mail-transport-agent]  3.2.2-1

Versions of packages popularity-contest suggests:
ii  anacron  2.3-24

-- debconf information excluded

Bug#865708: dh_installwm: the man page does not match to the code

[Robert Luberda]

Package: debhelper
Version: 10.2.5
Severity: minor

dh_installwm(1) says that the slave man page is optional:
   The window manager's man page is also  registered as a slave symlink (in v6 
mode 
   and up), if it is found in usr/share/man/man1/ in the package  build 
directory.
   

However dh_installwm's code does not entirely agree with that:

  if (! compat(9)) {
error("no manpage found (creating an x-window-manager alternative requires 
a slave symlink for the manpage)");

^
  }

Regards,
robert
 
-- System Information:
Debian Release: 9.0
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages debhelper depends on:
ii  autotools-dev20161112.1
ii  binutils 2.28-6
ii  dh-autoreconf14
ii  dh-strip-nondeterminism  0.035-2
ii  dpkg 1.18.24
ii  dpkg-dev 1.18.24
ii  file 1:5.30-1
ii  libdpkg-perl 1.18.24
ii  man-db   2.7.6.1-2
ii  perl 5.24.1-4
ii  po-debconf   1.0.20

debhelper recommends no packages.

Versions of packages debhelper suggests:
pn  dh-make  

-- no debconf information

Bug#865707: debhelper: Please support DEB_BUILD_OPTIONS=nodoc

[Robert Luberda]

Package: debhelper
Version: 10.2.5
Severity: wishlist

The Debian Policy has recently introduced DEB_BUILD_OPTIONS=nodoc. 
I've just tried to handle it in one of my packages, namely afterstep,
but it quickly turned out that if the documentation files are not
generated, then dh_install fails because it still trying to install 
them. Fortunatelly my debian/*.install and other debhelper files are 
generated, so I can easily filter /usr/share/{doc,man} out of them,
but then dh_installwm fails due to missing man page.

Could you please add support for DEB_BUILD_OPTIONS=nodoc into debhelper,
so that it will not try to install anything into /usr/share/doc (except
copyright, changelog, and maybe NEWS.Debian), and /usr/share/man? Also
it should not install doc-base files either for obvious reason.

I think this will require changes at least in dh_install, dh_installdoc,
dh_installman, dh_installwm.

(The Policy says in section 4.9.1 that the `nodoc' option should cause build 
process to skip generation of documentation files, like man pages, API 
documentation, etc. This probably means that the non-generated documentation 
can be sill installed, however I am not sure if debhelper will be able to 
distinguish between generated and non-generated doc files on its own, 
that's why I think that debhelper in the `nodoc' mode should simply avoid 
installing any documentation).


Regards,
robert

-- System Information:
Debian Release: 9.0
  APT prefers unstable
  APT policy: (990, 'unstable'), (990, 'testing'), (990, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8), 
LANGUAGE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages debhelper depends on:
ii  autotools-dev20161112.1
ii  binutils 2.28-6
ii  dh-autoreconf14
ii  dh-strip-nondeterminism  0.035-2
ii  dpkg 1.18.24
ii  dpkg-dev 1.18.24
ii  file 1:5.30-1
ii  libdpkg-perl 1.18.24
ii  man-db   2.7.6.1-2
ii  perl 5.24.1-4
ii  po-debconf   1.0.20

debhelper recommends no packages.

Versions of packages debhelper suggests:
pn  dh-make  

-- no debconf information

Bug#614204: */updates and *-updates are confusingly similar

[Robert Luberda]

On nie, 20 lut 2011, Thijs Kinkhorst wrote:

> On Sunday 20 February 2011 12:04:08 Julian Andres Klode wrote:
> > With squeeze, squeeze-updates was introduced, bringing
> > us in a situation where we have two repositories:
> > 
> > squeeze-updates (for updates)
> > squeeze/updates (for security)
> > 
> > That's confusing. It would be better to rename */updates
> > to *-security.
> 
> I'm not so sure this is such a problem, since you leave out essential 
> information. If you present it like this, and this is the way users will 
> configure it, the difference becomes instantly clear:
> 
> deb http://security.debian.org/ squeeze/updates main contrib non-free
> deb http://ftp.nl.debian.org/debian/ squeeze-updates main contrib non-free

Even though I agree this is not a big issue, it still seems to me 
a bit strange or at least inconsistent. IMHO it would look much 
better if the same separator, i.e. either '/' or '-', was used in 
both places. However doing this would most probably make it hard 
to setup local mirrors containig both suites together.

Regards,
robert

Bug#865222: ftp.debian.org: incorrect links in README.html

[Robert Luberda]

Package: ftp.debian.org
Severity: minor

The http://ftp.debian.org/debian/README.html page contains the following
paragraph about stretch:

  Debian 9.0, or stretch
Debian 9.0 was released Saturday, 17th June 2017. Installation and 
upgrading instructions, More information 

but both "Installation and upgrading instructions" [1] and "More information" 
[2] links point to
the jessie documentation, not to the stretch one.

[1] https://www.debian.org/releases/jessie/amd64/
[2] https://www.debian.org/releases/jessie/

Regards,
robert

Bug#864198: unblock: upx-ucl/3.91-4

[Robert Luberda]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Please unblock version 3.91-4 of package upx-ucl.

This version fixes crashes occurring while executing compressed i386 
position-indpendent executables, for example the following commands 
on i386 system:
  cp /usr/bin/vim .; upx ./vim; ./vim
  cp /usr/bin/mutt .; upx ./mutt; ./mutt
end with segmentation faults in stretch, while they work as expected 
in jessie. This is related to some changes in binutils output.  (Also 
note that vim is PIE in both jessie and stretch, while mutt wasn't
PIE in jessie).

This is the issue I wrote about a week ago in #863537 while requesting 
unblock for previous upload that addressed crashes of amd64 executables,
and it has been fixed in a pretty same way as the amd64 one.


Comparing upx-ucl_3.91-3.dsc upx-ucl_3.91-4.dsc
diff -Nru upx-ucl-3.91/debian/changelog upx-ucl-3.91/debian/changelog
--- upx-ucl-3.91/debian/changelog   2017-05-20 09:16:37.0 +0200
+++ upx-ucl-3.91/debian/changelog   2017-05-31 21:50:25.0 +0200
@@ -1,3 +1,12 @@
+upx-ucl (3.91-4) unstable; urgency=medium
+
+  * Introduce Check-DT_REL-i386.patch to fix segmentation faults occurring
+while executing upx-compressed i386 position-independent binaries. The
+patch is strongly based on the one added previously for amd64 binaries,
+and was reported to upstream in https://github.com/upx/upx/issues/106.
+
+ -- Robert Luberda <rob...@debian.org>  Wed, 31 May 2017 21:50:25 +0200
+
 upx-ucl (3.91-3) unstable; urgency=medium
 
   * Add Check-DT_RELA.patch based on upstream's commit d688a05ac7 to
diff -Nru upx-ucl-3.91/debian/patches/Check-DT_REL-i386.patch 
upx-ucl-3.91/debian/patches/Check-DT_REL-i386.patch
--- upx-ucl-3.91/debian/patches/Check-DT_REL-i386.patch 1970-01-01 
01:00:00.0 +0100
+++ upx-ucl-3.91/debian/patches/Check-DT_REL-i386.patch 2017-05-31 
21:50:25.0 +0200
@@ -0,0 +1,66 @@
+From: Robert Luberda <rob...@debian.org>
+Date: Sun, 28 May 2017 12:27:11 +0200
+Subject: DT_JMPREL is missing on i386 as well; check DT_REL
+
+Apply the changes from commit d688a05ac78517bcba09bae0f60bc76f3aa51ddb
+to PackLinuxElf32::canPack(), however check DT_REL instread of DT_RELA.
+This fixes crashes while running position independent i386 executables
+compressed with upx.
+
+The patch was sent to upstream in https://github.com/upx/upx/issues/106,
+and eventually got applied in a sligthly modified form in commit
+https://github.com/upx/upx/commit/ee18fe9bbab4955191e68a0982196f1b6f3e5c7d
+(the differences include extractions of duplicated code into functions,
+and introduction of DT_1_PIE checks that depend on  code not yet
+available in upx 3.91).
+---
+ src/p_elf_enum.h |  1 +
+ src/p_lx_elf.cpp | 16 
+ 2 files changed, 17 insertions(+)
+
+diff --git a/src/p_elf_enum.h b/src/p_elf_enum.h
+index 5a4f602..121d04d 100644
+--- a/src/p_elf_enum.h
 b/src/p_elf_enum.h
+@@ -150,6 +150,7 @@
+ DT_RELAENT  =  9,   /* Size of one RELA relocation */
+ DT_INIT = 12,   /* Address of init function */
+ DT_REL  = 17,   /* Relocations which contain no addend */
++DT_RELSZ   =  18,   /* Total size of Rel relocs */
+ DT_RELENT   = 19,   /* Size of one Rel relocation */
+ DT_STRSZ= 10,   /* Sizeof string table */
+ DT_PLTREL   = 20,   /* Type of reloc in PLT */
+diff --git a/src/p_lx_elf.cpp b/src/p_lx_elf.cpp
+index 6f95c0b..ab00666 100644
+--- a/src/p_lx_elf.cpp
 b/src/p_lx_elf.cpp
+@@ -1359,6 +1359,8 @@ bool PackLinuxElf32::canPack()
+ // defined symbols, and there might be no DT_HASH.
+ 
+ Elf32_Rel const *
++rel= (Elf32_Rel const *)elf_find_dynamic(Elf32_Dyn::DT_REL);
++Elf32_Rel const *
+ jmprel= (Elf32_Rel const *)elf_find_dynamic(Elf32_Dyn::DT_JMPREL);
+ for (   int sz = elf_unsigned_dynamic(Elf32_Dyn::DT_PLTRELSZ);
+ 0 < sz;
+@@ -1372,6 +1374,20 @@ bool PackLinuxElf32::canPack()
+ goto proceed;
+ }
+ 
++// 2017-05-28 DT_JMPREL is no more (binutils-2.26.1)?
++// Check the general case, too.
++for (   int sz = elf_unsigned_dynamic(Elf32_Dyn::DT_RELSZ);
++0 < sz;
++(sz -= sizeof(Elf32_Rel)), ++rel
++) {
++unsigned const symnum = get_te32(>r_info) >> 8;
++char const *const symnam = get_te32([symnum].st_name) + 
dynstr;
++if (0==strcmp(symnam, "__libc_start_main")
++||  0==strcmp(symnam, "__uClibc_main")
++||  0==strcmp(symnam, "__uClibc_start_main"))
++goto proceed;
++}
++
+ // Heuristic HACK for shared libraries (compare Darwin (MacOS) Dylib.)
+ // If there is an existing DT_INIT, and if everything that the dynamic
+ // linker ld-linux needs to

Bug#863896: dnsmasq: systemd appends junk to the dnsmasq commandline

[Robert Luberda]

Control: fixed -1 2.77-1

> 
> This issue seems to be caused by the recent upgrade of the dns-root-data
> package. The root.ds file in this new package contains two records where
> the previous version (2015052300+h+1) only had one record.

It looks like the bug has been recently fixed in unstable.

Regards,
robert

Bug#864016: dns-root-data: Upgrade breaks dnsmasq

[Robert Luberda]

Package: dns-root-data
Version: 2017041101
Severity: serious

Upgraded dns-root-data should declare "Breaks: dnsmasq (<< 2.77-1~)",
see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=863896#15

Regards,
robert


-- System Information:
Debian Release: 9.0
  APT prefers testing
  APT policy: (990, 'testing'), (200, 'unstable')
Architecture: amd64
 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

-- no debconf information

Bug#863537: unblock: upx-ucl/3.91-3

[Robert Luberda]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Please unblock version 3.91-3 of package upx-ucl

It contains a patch taken from upstream to fix #860953, i.e. crashes
of binaries compressed with upx on amd64, for example:

  robert@vox:~$ cp /usr/bin/mutt .; upx mutt 
 Ultimate Packer for eXecutables
Copyright (C) 1996 - 2013
  UPX 3.91Markus Oberhumer, Laszlo Molnar & John Reiser   Sep 30th 2013
  
  File size Ratio  Format  Name
    --   ---   ---
 1178240 ->623212   52.89%  linux/ElfAMD   mutt 
 
  
  Packed 1 file.
  robert@vox:~$ ./mutt
  Segmentation fault (core dumped)

The crashes are related to some changes in binutils (most probably) 2.26.1,
and this is a regression against jessie: I've just checked that
compressing with upx 3.91-2 and then executing mutt from 1.6.2-3 amd64 package 
[1] 
works correctly, while doing the same with mutt from 1.7.0-1 [2] crashes.

[1] 
http://snapshot.debian.org/archive/debian/20160824T042609Z/pool/main/m/mutt/mutt_1.6.2-3_amd64.deb
[2] 
http://snapshot.debian.org/archive/debian/20160829T100430Z/pool/main/m/mutt/mutt_1.7.0-1_amd64.deb

Also I've just discovered that we may need yet another upload probably with a
pretty similar patch to fix similar issue on i386, as mutt:i386 compressed 
with upx-ucl:i386 (even 3.94-1 from experimental) crashes as well :(. (I
did the same check a week ago, but using `ls' binary, and it didn't
crashed, so I thought upx-ucl:i386 worked). I will try to work with 
upstream on a fix for i386.

The version 3.91-3 contains also two small unrelated fixes:
 - simple fix for FTBFS on sparc64
 - update upstream homepage URL and debian/watch
I hope they are OK for you, but I can revert them via another upload if not.


Comparing upx-ucl_3.91-2.dsc upx-ucl_3.91-3.dsc
diff -Nru upx-ucl-3.91/debian/changelog upx-ucl-3.91/debian/changelog
--- upx-ucl-3.91/debian/changelog   2016-07-03 13:02:59.0 +0200
+++ upx-ucl-3.91/debian/changelog   2017-05-20 09:16:37.0 +0200
@@ -1,3 +1,15 @@
+upx-ucl (3.91-3) unstable; urgency=medium
+
+  * Add Check-DT_RELA.patch based on upstream's commit d688a05ac7 to
+fix segmentation faults on decompressing executables produced by
+newer binutils on amd64 (closes: #860953).
+  * Apply the following packaging changes from 3.94-1 in Debian/experimental:
++ upstream homepage was moved to github, update URLs in debian/watch,
+  debian/control and debian/copyright;
++ debian/rules: Add sparc64 to big-endian archs to fix FTBFS.
+
+ -- Robert Luberda <rob...@debian.org>  Sat, 20 May 2017 09:17:12 +0200
+
 upx-ucl (3.91-2) unstable; urgency=medium
 
   * Add 03-FTBFS-GCC6.patch to fix build failure with g++-6 (closes: #811595).
diff -Nru upx-ucl-3.91/debian/control upx-ucl-3.91/debian/control
--- upx-ucl-3.91/debian/control 2016-07-03 13:02:59.0 +0200
+++ upx-ucl-3.91/debian/control 2017-05-20 09:16:37.0 +0200
@@ -7,7 +7,7 @@
libucl-dev (>= 1.03),
lzma-dev (>= 9.22),
zlib1g-dev (>= 1:1.1.4)
-Homepage: http://upx.sourceforge.net/
+Homepage: https://upx.github.io/
 Vcs-Git: https://anonscm.debian.org/git/users/robert/upx-ucl.git
 Vcs-Browser: https://anonscm.debian.org/cgit/users/robert/upx-ucl.git
 
diff -Nru upx-ucl-3.91/debian/copyright upx-ucl-3.91/debian/copyright
--- upx-ucl-3.91/debian/copyright   2016-07-03 13:02:59.0 +0200
+++ upx-ucl-3.91/debian/copyright   2017-05-20 09:16:37.0 +0200
@@ -1,6 +1,6 @@
 Format: https://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
 Upstream-Contact: Markus F.X.J Oberhumer <mar...@oberhumer.com>
-Source: http://upx.sourceforge.net/
+Source: https://upx.github.io/
 
 Files: *
 Copyright: 1996-2013, Markus Franz Xaver Johannes Oberhumer
diff -Nru upx-ucl-3.91/debian/patches/Check-DT_RELA.patch 
upx-ucl-3.91/debian/patches/Check-DT_RELA.patch
--- upx-ucl-3.91/debian/patches/Check-DT_RELA.patch 1970-01-01 
01:00:00.0 +0100
+++ upx-ucl-3.91/debian/patches/Check-DT_RELA.patch 2017-05-20 
09:16:37.0 +0200
@@ -0,0 +1,61 @@
+From: John Reiser <jrei...@bitwagon.com>
+Date: Sun, 9 Oct 2016 21:50:45 -0700
+Subject: DT_JMPREL vanished (binutils-2.26.1); check all DT_RELA.
+
+   modified:   p_elf_enum.h
+   modified:   p_lx_elf.cpp
+
+[ Patch taken from 
https://github.com/upx/upx/commit/d688a05ac78517bcba09bae0f60bc76f3aa51ddb ]
+
+Bugs-Debian: https://bugs.debian.org/#860953
+---
+ src/p_elf_enum.h |  1 +
+ src/p_lx_elf.cpp | 16 
+ 2 files changed, 17 insertions(+)
+
+diff --git a/src/p_elf_enum.h b/src/p_elf_enum.h
+index 52093ee..5a4f602 100644
+--- a/src/p_elf_enum.h
 b/src/p_elf_enum.h
+@@ -146,6 +146,7 @@
+ DT_STRTAB   =  5,   /* S

Bug#863065: fonty-rg: Recommends removed package, and contains unusable scripts

[Robert Luberda]

Package: fonty-rg
Version: 0.6
Severity: serious
Justification: Policy 3.6

The 'utf8' and 'iso' scripts included in fonty-rg use 'consolechars'
command which is not provided by any Debian package, even in jessie, as
the 'console-tools' package was removed from Debian in 2013 (see #671342).
Nevertheless in 2015 fonty-rg started recommending the removed package 
as a fix for #487514.  (Another, unreleated, issue is that the suggested 
'fonty' package is not in Debian either, see #474125).

In my opinion both 'utf8' and 'iso' scripts should be:
 - either removed together with their manpages (or maybe replaced 
   with some documentation in README.Debian explaining how to enable
   fonts from the package?),
 - or rewritten to use console utilities from the kbd package.

Regards,
robert

-- System Information:
Debian Release: 9.0
  APT prefers testing
  APT policy: (990, 'testing'), (200, 'unstable')
Architecture: amd64
 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

fonty-rg depends on no packages.

Versions of packages fonty-rg recommends:
pn  console-tools  

Versions of packages fonty-rg suggests:
pn  fonty  

-- no debconf information

Bug#862363: jessie-pu: package dwww/1.12.1+deb8u1

[Robert Luberda]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Adrian Bunk wrote:

Hi,

> 
> debian/changelog |8  scripts/dwww-convert |2
> +- 2 files changed, 9 insertions(+), 1 deletion(-)
> 
> diff -Nru dwww-1.12.1/debian/changelog
> dwww-1.12.1+deb8u1/debian/changelog ---
> dwww-1.12.1/debian/changelog  2014-01-14 00:10:17.0 +0200 
> +++ dwww-1.12.1+deb8u1/debian/changelog   2017-05-11
> 22:14:30.0 +0300 @@ -1,3 +1,11 @@ +dwww (1.12.1+deb8u1)
> jessie; urgency=medium + +  * Non-maintainer upload. +  * Apply fix
> from Holger Spielmann for the `Last-Modified' header. +(Closes:
> #850016, #850885) + + -- Adrian Bunk   Thu, 11 May
> 2017 22:07:36 +0300 + dwww (1.12.1) unstable; urgency=medium
> 
> * apache.conf: add an `Alias /dwww /var/www/dwww' line to
> accommodate diff -Nru dwww-1.12.1/scripts/dwww-convert
> dwww-1.12.1+deb8u1/scripts/dwww-convert ---
> dwww-1.12.1/scripts/dwww-convert  2014-01-14 00:10:17.0
> +0200 +++ dwww-1.12.1+deb8u1/scripts/dwww-convert 2017-05-11
> 22:03:26.0 +0300 @@ -327,7 +327,7 @@ print "Content-type:
> $mime_type" . (defined $mime_charset ? "; charset=$mime_charset\n"
> : "\n"); my @stat = stat( $filename ); my $mtime = $stat[9]; -
> print "Last modified: " . gmtime($mtime) . "\n"; +print
> "Last-modified: " . gmtime($mtime) . "\n"; print
> "Content-Disposition: inline; filename=\"$base_name\"\n"; print
> "\n"; } # }}}
> 


Release Team,

Could you please approve this change and allow Adrian to proceed with
the NMU?

It fixes a pretty old bug in dwww that was recently made visible (and
 thus made dwww mostly unusable) due to the security upload of apache2
into jessie.

Regads,
robert, dwww maintainer



-BEGIN PGP SIGNATURE-
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=zpV9
-END PGP SIGNATURE-

Bug#860953: upx-ucl: compressed executable segfaults and cannot be decompressed

[Robert Luberda]

b...@debian.org wrote:
Hi,

> 
> Packed 1 file.
> 
> $ ./freedink --help
> Segmentation fault

Thanks for the report. I've discovered today that the bug  affects more
binaries on amd64, for example /bin/ls, but not all of them, .e.g
/bin/bash seems to be unaffected. Could you please confirm this on your
system?

> 
> Note: this issue does not appear in experimental's 3.93.

After some play with git bisect I found the commit that has fixed it:
https://github.com/upx/upx/commit/d688a05ac78517bcba09bae0f60bc76f3aa51ddb

Regards,
robert

Bug#861381: unblock: man2html/1.6g-9

[Robert Luberda]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Please unblock package man2html/1.6g-9

It fixes a bug in parsing of tbl tables that causes man2html to generate
so huge output when called on printf(1posix) man page that it appears to
to hang in an endless loop. The bug was reported to Ubuntu, see 
https://bugs.launchpad.net/ubuntu/+source/man2html/+bug/1666067 .

I discovered that the main cause of the issue lies in the following code 
that parses tbl headers:

   while (*c && *c!='.') {
 switch (*c) {
 // stripped 
  case 'f': case 'F':
c++;
curfield->font = toupper(*c);
c++;
if (!isspace(*c)) c++;
break;
 // stripped

The last two increments of the `c' variable ignore the fact that the
variable can be equal to '.' that should cause the loop to break. 
The '.' character means the end of the tbl header, but because of the
bug man2html didn't notice that the header ended, it interpreted numbers
like  that existed in the tbl body as widths of table's
columns and then tried to apply such a huge widths by displayng `'
strings...

To fix the issue I made the increments conditional by adding 
"if (*c != '.')" checks. (When I think about it now, I realize 
that I should have checked for *c != '\0' as well just in case).

I also run old and new versions of man2html on all manpages installed on
my system, and it looks like the bug impacted a few more pages (but
without making man2html to appear to hng on them) from manpages-posix 
package (namely {awk,ed,expr,gencat,lex,uunencode}(1posix), and additionally
ethtool(8)), so its impact is not really big, but on the other hand
the fix is really small, that's why I'd like to ask you to consider
unblocking the package. 

Thanks,
robert

diff -Nru man2html-1.6g/debian/changelog man2html-1.6g/debian/changelog
--- man2html-1.6g/debian/changelog  2015-10-26 00:58:23.0 +0100
+++ man2html-1.6g/debian/changelog  2017-04-19 22:54:48.0 +0200
@@ -1,3 +1,15 @@
+man2html (1.6g-9) unstable; urgency=medium
+
+  * Add 036-fix-tbl-font-parsing.patch not to ignore the tbl header end
+while parsing tbl font specifiers what caused man2html to hang on
+printf(1posix) man page (LP: #1666067)
+  * man2html.cgi.c: fix a small typo found by lintian.
+  * debian/copyright:
++ add Vcs-Git and Vcs-Browser fields;
++ bump Standards-Version to 3.9.8 (no changes).
+
+ -- Robert Luberda <rob...@debian.org>  Wed, 19 Apr 2017 22:54:48 +0200
+
 man2html (1.6g-8) unstable; urgency=medium
 
   * Add 035-source-date-epoch.patch to produce reproducible timestamps in
diff -Nru man2html-1.6g/debian/control man2html-1.6g/debian/control
--- man2html-1.6g/debian/control2015-10-26 00:58:23.0 +0100
+++ man2html-1.6g/debian/control2017-04-19 22:54:48.0 +0200
@@ -3,8 +3,10 @@
 Priority: optional
 Maintainer: Robert Luberda <rob...@debian.org>
 Build-Depends: debhelper (>= 9)
-Standards-Version: 3.9.6
+Standards-Version: 3.9.8
 Homepage: http://users.actrix.gen.nz/michael/vhman2html.html
+Vcs-Git: https://anonscm.debian.org/git/users/robert/man2html.git
+Vcs-Browser: https://anonscm.debian.org/cgit/users/robert/man2html.git
 
 Package: man2html-base
 Architecture: any
diff -Nru man2html-1.6g/debian/patches/036-fix-tbl-font-parsing.patch 
man2html-1.6g/debian/patches/036-fix-tbl-font-parsing.patch
--- man2html-1.6g/debian/patches/036-fix-tbl-font-parsing.patch 1970-01-01 
01:00:00.0 +0100
+++ man2html-1.6g/debian/patches/036-fix-tbl-font-parsing.patch 2017-04-19 
22:54:48.0 +0200
@@ -0,0 +1,42 @@
+From: Robert Luberda <rob...@debian.org>
+Date: Wed, 19 Apr 2017 22:12:46 +0200
+Subject: Fix tbl font parsing
+
+The printf.1posix man page contains tbl table that looks like this:
+.TS
+lf5 | lf5 | lf7.
+_
+99@2147483647@printf: "99" arithmetic overflow
+.TE
+
+While parsing the last 'f' from the header line man2html used to ignore
+all the characters until the next space. Unfortunately the included
+the '.' (dot) that ends the header. As a result man2html considered the
+table data lines as being part of the header and thus set the length
+of cells to 99 and 2147483647 characters. Finally it tried to
+display such an unreasonable number of  entities, what make it
+to appear to hang...
+
+To fix the issue make sure the trailing dot is not ignored while parsing
+'f' or 'F' specifiers of tbl header.
+
+Bugs-Ubuntu: https://bugs.launchpad.net/ubuntu/+source/man2html/+bug/1666067
+---
+ man2html/man2html.c | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/man2html/man2html.c b/man2html/man2html.c
+index 8c4db71..9da6d19 100644
+--- a/man2html/man2html.c
 b/man2html/man2html.c
+@@ -902,8 +902,8 @@ static char *scan_format(char *c, TABLEROW **result, int 
*maxcol)
+   case 'f': case 'F':
+   

Bug#861020: devscripts: debuild should not sign UNRELEASED builds

[Robert Luberda]

Package: devscripts
Version: 2.17.5
Severity: wishlist


Currently while building an UNRELEASED package with dpkg-buildpackage, it
skips signing the package:
dpkg-buildpackage: warning: not signing UNRELEASED build; use --force-sign to 
override

Could you please do the same in debuild?

Thanks,
robert

-- System Information:
Debian Release: 9.0
  APT prefers testing
  APT policy: (990, 'testing'), (200, 'unstable')
Architecture: amd64
 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages devscripts depends on:
ii  dpkg-dev 1.18.23
ii  libc62.24-9
ii  perl 5.24.1-2
pn  python3:any  

Versions of packages devscripts recommends:
ii  apt 1.4
pn  at  
ii  dctrl-tools 2.24-2+b1
ii  debian-keyring  2017.01.20
ii  dput0.12.0
pn  equivs  
ii  fakeroot1.21-3.1
ii  file1:5.29-3
ii  gnupg   2.1.18-6
ii  gnupg2  2.1.18-6
pn  libdistro-info-perl 
ii  libdpkg-perl1.18.23
ii  libencode-locale-perl   1.05-1
pn  libgit-wrapper-perl 
pn  liblist-compare-perl
ii  liblwp-protocol-https-perl  6.06-2
pn  libsoap-lite-perl   
ii  liburi-perl 1.71-1
ii  libwww-perl 6.15-1
ii  licensecheck3.0.29-1
ii  lintian 2.5.50.1
ii  man-db  2.7.6.1-2
ii  patch   2.7.5-1+b2
ii  patchutils  0.3.4-2
ii  python3-debian  0.1.30
pn  python3-magic   
ii  sensible-utils  0.0.9
ii  strace  4.15-2
ii  unzip   6.0-21
ii  wdiff   1.2.2-2
ii  wget1.18-5
ii  xz-utils5.2.2-1.2+b1

Versions of packages devscripts suggests:
pn  adequate 
pn  autopkgtest  
pn  bls-standalone   
ii  bsd-mailx [mailx]8.1.2-0.20160123cvs-4
ii  build-essential  12.3
pn  check-all-the-things 
pn  cvs-buildpackage 
pn  devscripts-el
pn  diffoscope   
pn  disorderfs   
pn  dose-extra   
pn  duck 
pn  faketime 
pn  gnuplot  
ii  gpgv 2.1.18-6
pn  how-can-i-help   
ii  libauthen-sasl-perl  2.1600-1
ii  libfile-desktopentry-perl0.22-1
pn  libnet-smtps-perl
pn  libterm-size-perl
ii  libtimedate-perl 2.3000-2
pn  libyaml-syck-perl
ii  mailutils [mailx]1:3.1.1-1
pn  mozilla-devscripts   
ii  mutt 1.7.2-1
ii  openssh-client [ssh-client]  1:7.4p1-10
pn  piuparts 
pn  ratt 
pn  reprotest
pn  svn-buildpackage 
pn  w3m  

-- no debconf information

Bug#860371: unblock: sysstat/11.4.3-1

[Robert Luberda]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

Pleasese consider unblocking version 11.4.3-1 of package sysstat.

This is the latest version of sysstat considered as stable by upstream,
and it has been in unstable for over last two months without new bugs
reported.

I can see Ubuntu also has that version since that time. They have one
new issue, https://bugs.launchpad.net/ubuntu/+source/sysstat/+bug/1680705,
but I think this is something that was actually fixed in 11.4.3 rather
than introduced (unfortunatelly the Launchpad interface does not
mention the version the bug refers to), because it looks like this is
the same bug as https://bugs.debian.org/842831 that was fixed by upstream 
in 11.4.3 (actually 11.4.2, which wasn't uploaded to Debian).

The debdiff is attached.

unblock sysstat/11.4.3-1

Thanks,
robert

-- System Information:
Debian Release: 9.0
  APT prefers testing
  APT policy: (990, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=pl_PL.UTF-8, LC_CTYPE=pl_PL.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)


sysstat-11.4.3.diff.gz
Description: application/gzip