[Touch-packages] [Bug 1923845] Re: Please compress packages with zstd by default

2021-06-10 Thread Julian Andres Klode
** Changed in: apt (Ubuntu)
   Status: New => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923845

Title:
  Please compress packages with zstd by default

Status in appstream-glib package in Ubuntu:
  New
Status in apt package in Ubuntu:
  Fix Released
Status in aptly package in Ubuntu:
  New
Status in boinc package in Ubuntu:
  New
Status in busybox package in Ubuntu:
  New
Status in cdebootstrap package in Ubuntu:
  New
Status in cdist package in Ubuntu:
  New
Status in debdelta package in Ubuntu:
  New
Status in debian-el package in Ubuntu:
  New
Status in debootstrap package in Ubuntu:
  Fix Released
Status in debsig-verify package in Ubuntu:
  New
Status in debsigs package in Ubuntu:
  New
Status in diffoscope package in Ubuntu:
  New
Status in dpkg package in Ubuntu:
  New
Status in dpkg-sig package in Ubuntu:
  New
Status in file package in Ubuntu:
  New
Status in libsolv package in Ubuntu:
  New
Status in lintian package in Ubuntu:
  New
Status in lutris package in Ubuntu:
  New
Status in obs-build package in Ubuntu:
  New
Status in osc package in Ubuntu:
  New
Status in radare2 package in Ubuntu:
  New
Status in reprepro package in Ubuntu:
  New
Status in vim-scripts package in Ubuntu:
  New
Status in zeroinstall-injector package in Ubuntu:
  New

Bug description:
  https://people.canonical.com/~rbalint/zstd-debs/ contains a .deb built
  on Hirsute having both data and control members of the .deb being
  compressed with zstd. It can be handy for testing various tools.

  [dpkg]
  Decompression support in dpkg landed first in Bionic and is being SRUd to 
Xenial in LP: #1764220 enable Launchpad's Xenial systems to process the 
zstd-compressed binary packages.
  From dpkg's perspective the upgrade path is cleared.

  The original plan was compressing only the internal data.tar .deb
  member, but dpkg uses uniform compression by default since dpkg 1.19.0
  thus I'm collecting all the changes to support control.tar.zst, too,
  in this bug.

  Reviewed packages from:
  https://codesearch.debian.net/search?q=data.tar.xz=1=1
  https://codesearch.debian.net/search?q=control.tar.xz=1=1

  appstream-glib  - needs fix: libappstream-builder/asb-package-deb.c
  aptly   - needs fix: deb/deb.go
  boinc   - needs fix: debian/fetch_example_applications.sh
  busybox - needs fix: archival/dpkg_deb.c archival/dpkg.c
  cdebootstrap- needs fix: src/package.c
  cdist   - may need fix, can use dpkg-deb: 
cdist/preos/debootstrap/files/devuan-debootstrap/functions
  debdelta- needs fix: debdelta debpatch.sh
  debian-el   - needs fix: deb-view.el
  debian-handbook - needs fix, maybe later, for Debian
  debootstrap - needs fix, 
https://salsa.debian.org/installer-team/debootstrap/-/merge_requests/54
  debsigs - needs fix, debsigs
  debsig-verify   - needs fix, src/debsig-verify.c
  diffoscope  - needs fix, diffoscope/comparators/deb.py
  dpkg- needs fix, change default
  dpkg-sig- needs fix, dpkg-sig
  dpmb- needs fix, maybe later, for Debian
  elfutils- may need fix, uses dpkg-deb if it is available, does not 
handle .gz either
  file- needs fix, magic/Magdir/archive
  libsolv - needs fix, ext/repo_deb.c
  lintian - needs fix malformed-deb-archive
  lutris  - needs fix, lutris/util/extract.py
  obs-build   - needs fix Build/Deb.pm
  osc - needs fix osc/util/debquery.py control.tar.zst only
  python-apt  - needs fix 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()
  radare2 - needs fix
  reprepro- needs fix, debfile.c
  vim-scripts - needs fix debPlugin/autoload/deb.vim
  winetricks  - needs fix when Debian switches src/winetricks
  zeroinstall-injector - needs fix src/zeroinstall/archive.ml

  acr - skip, does not _have to_ be fixed, just creates packages, 
see dist/deb_hand.mak
  alien   - skip, uses dpkg-deb to extract .deb
  ansible - not affected, just test data in dbdata.tar.xz
  anthy   - not affected, just changelog entry
  apt - seems fixed already
  ceph- not affected in Ubuntu's version
  circlator   - not affected, just test data
  cowdancer   - not affected, just documentation
  eccodes - skip, just orig-data.tar.xz
  eckit   - skip, just ...orig-data.tar.xz
  firefox - skip, profdata.tar.xz
  firefox-esr - skip, profdata.tar.xz
  galculator  - skip, just changelog
  grads   - skip, ...orig-data.tar.xz
  gvmd- skip, just creates xz compressed .deb
  insighttoolkit4 - skip, ...orig-data.tar.xz
  jdeb- skip, just creates compressed .deb packages
  jmol- skip, just local data archive
  jq  - skip, 

[Touch-packages] [Bug 1931178] Re: apt-key fails when POSIXLY_CORRECT environmental variable is set

2021-06-08 Thread Julian Andres Klode
Is this a regression from an older release? We can fix it, but if it's
not a regression, I'd just skip it, as running apt-key is the wrong
approach anyway.

They should just pmrum cp some-repo-key.gpg /etc/apt/trusted.gpg.d.
optimally drop the keyring elsewhere and use signed-by in sources.list
snippet, but oh well, it does not add much security really.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1931178

Title:
  apt-key fails when POSIXLY_CORRECT environmental variable is set

Status in apt package in Ubuntu:
  New

Bug description:
  apt-key fails when the POSIXLY_CORRECT environmental variable is set.
  This is due to chmod acting differently with POSIXLY_CORRECT set: chmod tries 
to use `--` as a parameter with this variable.

  This prevents some third-party utilities, like pmrun, from using apt-
  key.

  ---

  chmod with POSIXLY_CORRECT unset succeeds with the following commands:
  ```
  touch testfile
  chmod 0644 -- testfile
  echo $?
  ```
  The final command prints "0".
  (This is what the apt-key script expects to happen.)

  ---

  chmod with POSIXLY_CORRECT set fails with the following commands:
  ```
  export POSIXLY_CORRECT=
  touch testfile
  chmod 0644 -- testfile
  echo $?
  ```
  The final two commands print these two lines, respectively:
  """
  chmod: cannot access '--': No such file or directory
  1
  """
  (This is unexpected for the apt-key script.)

  ---

  As such, when a third party utility that sets the POSIXLY_CORRECT 
environmental variable, like pmrun, tries to run apt-key, it will fail.
  Example use similar to what the reporting customer tried: `cat 
some-repo-key.gpg | pmrun apt-key add -`

  The chmod examples are run on an up-to-date Bionic container, 
/usr/bin/apt-key has `chmod  -- path` commands at line 338 and line 499:
  GNU coreutils package version 8.28-1ubuntu1
  apt package version 1.6.13

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1931178/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1930584] Re: InstallProgress is not being used as a context manager

2021-06-02 Thread Julian Andres Klode
I kind of don't need the bug in both places. I'd suggest to go fix your
backend to pass an install progress and use a with statement, that we
don't use one inside apt.cache was intentional.

This means it produces warnings if you don't pass one, but frankly,
that's bad practice.

** Changed in: python-apt (Ubuntu)
   Status: New => Won't Fix

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to python-apt in Ubuntu.
https://bugs.launchpad.net/bugs/1930584

Title:
  InstallProgress is not being used as a context manager

Status in python-apt package in Ubuntu:
  Won't Fix

Bug description:
  Package: python3-apt
  Version: 2.2.0

  As can be seen in the constructor, InstallProgress is designed to be
  used in a 'with' in order to properly close write_stream and
  status_stream:

  https://git.launchpad.net/python-apt/tree/apt/progress/base.py#n163

  This commit added support for using InstallProgress as a context
  manager:

  https://git.launchpad.net/python-
  apt/commit/?id=462c05b39eae5a0fa2270d329b03d4711742f20d

  It also modified a test, but didn't change the way InstallProgress is
  used by cache.py.

  This causes Python 3 to complain that InstallProgress's constructor
  leaves unclosed files:

  > /usr/lib/python3.9/glob.py:123: ResourceWarning: unclosed file 
<_io.TextIOWrapper name=4 mode='r' encoding='UTF-8'>
  >   with os.scandir(dirname) as it:
  > Object allocated at (most recent call last):
  >   File "/opt/thinlinc/modules/thinlinc/packageinstaller/__init__.py", 
lineno 33
  > OOO0o000 = globals ( ) [ IIi1i111IiII ] . Backend ( )
  >   File "/opt/thinlinc/modules/thinlinc/packageinstaller/aptbackend.py", 
lineno 88
  > self . _aptcallback = i1Ii1i ( self )
  >   File "/opt/thinlinc/modules/thinlinc/packageinstaller/aptbackend.py", 
lineno 31
  > super ( ) . __init__ ( )
  >   File "/usr/lib/python3/dist-packages/apt/progress/base.py", lineno 164
  > self.status_stream = os.fdopen(self.statusfd, "r")  # type: 
io.TextIOBase # noqa
  >   File "/usr/lib/python3.9/os.py", lineno 1023
  > return io.open(fd, *args, **kwargs)
  > /usr/lib/python3.9/glob.py:123: ResourceWarning: unclosed file 
<_io.TextIOWrapper name=5 mode='w' encoding='UTF-8'>
  >   with os.scandir(dirname) as it:
  > Object allocated at (most recent call last):
  >   File "/opt/thinlinc/modules/thinlinc/packageinstaller/__init__.py", 
lineno 33
  > OOO0o000 = globals ( ) [ IIi1i111IiII ] . Backend ( )
  >   File "/opt/thinlinc/modules/thinlinc/packageinstaller/aptbackend.py", 
lineno 88
  > self . _aptcallback = i1Ii1i ( self )
  >   File "/opt/thinlinc/modules/thinlinc/packageinstaller/aptbackend.py", 
lineno 31
  > super ( ) . __init__ ( )
  >   File "/usr/lib/python3/dist-packages/apt/progress/base.py", lineno 163
  > self.write_stream = os.fdopen(self.writefd, "w")  # type: io.TextIOBase
  >   File "/usr/lib/python3.9/os.py", lineno 1023
  > return io.open(fd, *args, **kwargs)

  
  The issue was identified on Ubuntu 21.04:

  > Linux ubuntu2104 5.11.0-17-generic #18-Ubuntu SMP Thu May 6 20:10:11
  > UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/python-apt/+bug/1930584/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 978587] Re: apt should ensure .deb are not corrupted before handing them to dpkg

2021-05-29 Thread Julian Andres Klode
** Changed in: apt (Ubuntu)
   Status: Confirmed => Invalid

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/978587

Title:
  apt should ensure .deb are not corrupted before handing them to dpkg

Status in apt package in Ubuntu:
  Invalid

Bug description:
  Upon upgrading to libreoffice-core 3.5.2 version, I stumbled upon what
  seems to be a bad download issue:

  Preparing to replace libreoffice-core 1:3.5.1-1ubuntu5 (using 
.../libreoffice-core_1%3a3.5.2-2ubuntu1_amd64.deb) ...
  rmdir: failed to remove `/var/lib/libreoffice/basis3.4/program/': No such 
file or directory
  rmdir: failed to remove `/var/lib/libreoffice/basis3.4': No such file or 
directory
  Unpacking replacement libreoffice-core ...
  dpkg-deb (subprocess): data: internal bzip2 read error: 'DATA_ERROR'
  dpkg-deb: error: subprocess  returned error exit status 2
  dpkg: error processing 
/var/cache/apt/archives/libreoffice-core_1%3a3.5.2-2ubuntu1_amd64.deb 
(--unpack):
   subprocess dpkg-deb --fsys-tarfile returned error exit status 2

  I was asked to file a bug about it, as it might be possible for dpkg
  to recover from that more gracefully.

  Further information upon requests.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/978587/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1886161] Re: Unattended upgrades should work on roaming laptops

2021-05-25 Thread Julian Andres Klode
@Robin You should be able to just run `systemctl start apt-daily.service
apt-daily-upgrade.service` from the hook. This will start those two
services - which run update / upgrade - in the same transaction, and
hence ordering will work correctly.

The timer will still run again, so it's not _optimal_, but oh well. An
optimal solution would just make the timer "catch up" when the
conditions apply. We'll be able to simulate that I guess by moving away
from ConditionACPower=true to checking battery state ourself, then
exiting with a special error code and retrying on that.

systemd at least was fixed to allow retrying oneshot units a while ago,
but we've not had the chance to rework the apt integration to do
retries.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1886161

Title:
  Unattended upgrades should work on roaming laptops

Status in apt package in Ubuntu:
  New
Status in gnome-settings-daemon package in Ubuntu:
  Invalid

Bug description:
  For many if not most laptop computers, Unattended Upgrades seems all
  but useless by design.

  For upgrades to happen, with default config, 3 conditions must be met:

  - an internet connection must be up and running when the timer or cron or 
anacron tries the unattended upgrade
  - the connection must not be metered, whatever that means 
(Skip-Updates-On-Metered-Connections "true")
  - the computer must be plugged in (OnlyOnACPower "true")

  These are insurmountable problems for many laptops on the go.
  Inevitably, security upgrades will almost never run unattended on such
  computers. I discovered with shock that Unattended Upgrades had almost
  never run on my laptop. I tried all possible config tweaks. Nothing
  worked reliably and in the end I gave up and wrote a upgrade script
  which uses Network Manager's connection-up hook. Unattended Upgrades
  needs to do something like this out of the box. Users should not need
  to write scripts to ensure security upgrades.

  Unattended upgrades is an excellent project for servers. But it really
  needs to work, out of the box, on laptops too.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1886161/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1925745] Re: Hirsute: Cannot install ubuntu-desktop due to unmet dependencies

2021-05-20 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Impish)
   Status: Confirmed => Fix Committed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1925745

Title:
  Hirsute: Cannot install ubuntu-desktop due to unmet dependencies

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Hirsute:
  Confirmed
Status in apt source package in Impish:
  Fix Committed

Bug description:
  I'm using the ubuntu-21.04-live-server-amd64 ISO to install my hirsute
  desktop machines so I can automate the installation.

  After installation, I'm trying to install the desktop via "apt install
  ubuntu-desktop".

  This fails with:
  "Some packages could not be installed. This may mean that you have
  requested an impossible situation or if you are using the unstable
  distribution that some required packages have not yet been created
  or been moved out of Incoming.
  The following information may help to resolve the situation:

  The following packages have unmet dependencies:
   ubuntu-release-upgrader-gtk : Depends: ubuntu-release-upgrader-core (= 
1:21.04.11) but 1:21.04.10 is to be installed
 Depends: python3-distupgrade (= 1:21.04.11) 
but 1:21.04.10 is to be installed
  E: Unable to correct problems, you have held broken packages."

  "apt policy ubuntu-release-upgrader-core" shows:
  "ubuntu-release-upgrader-core:
Installed: 1:21.04.10
Candidate: 1:21.04.10
Version table:
   1:21.04.11 1 (phased 20%)
  500 http://at.archive.ubuntu.com/ubuntu hirsute-updates/main amd64 
Packages
   *** 1:21.04.10 500
  500 http://at.archive.ubuntu.com/ubuntu hirsute/main amd64 Packages
  100 /var/lib/dpkg/status"

  
  I could not find any information what "phased 20%" means. It's obviously a 
major problem not being able to install the desktop on a fresh installation.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1925745/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1928687] Re: Avoid infinite loop on EOF on media change prompt

2021-05-17 Thread Julian Andres Klode
** Also affects: apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Impish)
   Importance: Undecided
   Status: New

** Changed in: apt (Ubuntu Impish)
   Status: New => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1928687

Title:
  Avoid infinite loop on EOF on media change prompt

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Hirsute:
  New
Status in apt source package in Impish:
  Fix Released

Bug description:
  [Impact]
  The cdrom media change prompt hangs on EOF. This causes the autopkgtest to 
hang when run in a debian lxd container for unknown reasons related to 
lxd/systemd isolation likely (it works fine in Ubuntu), but it's also 
surprising behavior in general.

  [Test plan]
  We did not really come up with a test for this except running autopkgtest -- 
lxd autopkgtest/debian/$suite/$arch. That being said, the change is minimal, 
and we do test the positive case in the test suite.

  [Where problems could occur]
  This change is isolated to the handling of pressing 'c' or hitting EOF inside 
the media change prompt, so any regressions would occur solely within that 
space. We do not foresee any regressions from the change as intended, as it 
seems unlikely that "hanging" is what you actually want.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1928687/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1928687] [NEW] Avoid infinite loop on EOF on media change prompt

2021-05-17 Thread Julian Andres Klode
Public bug reported:

[Impact]
The cdrom media change prompt hangs on EOF. This causes the autopkgtest to hang 
when run in a debian lxd container for unknown reasons related to lxd/systemd 
isolation likely (it works fine in Ubuntu), but it's also surprising behavior 
in general.

[Test plan]
We did not really come up with a test for this except running autopkgtest -- 
lxd autopkgtest/debian/$suite/$arch. That being said, the change is minimal, 
and we do test the positive case in the test suite.

[Where problems could occur]
This change is isolated to the handling of pressing 'c' or hitting EOF inside 
the media change prompt, so any regressions would occur solely within that 
space. We do not foresee any regressions from the change as intended, as it 
seems unlikely that "hanging" is what you actually want.

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1928687

Title:
  Avoid infinite loop on EOF on media change prompt

Status in apt package in Ubuntu:
  New

Bug description:
  [Impact]
  The cdrom media change prompt hangs on EOF. This causes the autopkgtest to 
hang when run in a debian lxd container for unknown reasons related to 
lxd/systemd isolation likely (it works fine in Ubuntu), but it's also 
surprising behavior in general.

  [Test plan]
  We did not really come up with a test for this except running autopkgtest -- 
lxd autopkgtest/debian/$suite/$arch. That being said, the change is minimal, 
and we do test the positive case in the test suite.

  [Where problems could occur]
  This change is isolated to the handling of pressing 'c' or hitting EOF inside 
the media change prompt, so any regressions would occur solely within that 
space. We do not foresee any regressions from the change as intended, as it 
seems unlikely that "hanging" is what you actually want.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1928687/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1928100] [NEW] TLS handshake is fatal, not transient - hence not retried

2021-05-11 Thread Julian Andres Klode
Public bug reported:

[Impact]
TLS handshakes are reported as fatal errors rather than transient errors like 
other connection errors on unencrypted connections. This seems wrong - a server 
may just be spinning up or down and fail during the handshake, and prevents 
retrying such downloads

[Test plan]

A test case has been added to the integration test suite that ensures
handshake errors are retried upon configuration.

[Where problems could occur]
Transient errors do not cause an error exit from apt update, so scripts relying 
on errors from TLS handshakes in such situations will fail.

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1928100

Title:
  TLS handshake is fatal, not transient - hence not retried

Status in apt package in Ubuntu:
  New

Bug description:
  [Impact]
  TLS handshakes are reported as fatal errors rather than transient errors like 
other connection errors on unencrypted connections. This seems wrong - a server 
may just be spinning up or down and fail during the handshake, and prevents 
retrying such downloads

  [Test plan]

  A test case has been added to the integration test suite that ensures
  handshake errors are retried upon configuration.

  [Where problems could occur]
  Transient errors do not cause an error exit from apt update, so scripts 
relying on errors from TLS handshakes in such situations will fail.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1928100/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1928088] [NEW] Empty pockets don't end up in the cache - E: The value 'hirsute-backports' is invalid for APT::Default-Release as such a release is not available in the sources

2021-05-11 Thread Julian Andres Klode
Public bug reported:

[Impact]
If a pocket is empty, apt will not download the empty packages files, causing 
the repository to not be added to the cache. This in turn breaks commands such 
as using apt install -t hirsute-backports as apt is unable to find the release 
in the archive.

[Test plan]
Assuming an empty pocket is around, use apt install -t - to 
see it's in the cache.

An appropriate test case for empty stuff should be added to the test
suite as well (FIXME before SRUing), the test suite is extensive and
prevents most regressions

[Where problems could occur]
TBD

[Other info]
Originally reported at 
https://www.reddit.com/r/Kubuntu/comments/n9vg9e/cant_make_backports_repo_for_hirsute_to_work/

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: New

** Description changed:

  [Impact]
  If a pocket is empty, apt will not download the empty packages files, causing 
the repository to not be added to the cache. This in turn breaks commands such 
as using apt install -t hirsute-backports as apt is unable to find the release 
in the archive.
  
  [Test plan]
  Assuming an empty pocket is around, use apt install -t - to 
see it's in the cache.
  
  An appropriate test case for empty stuff should be added to the test
  suite as well (FIXME before SRUing), the test suite is extensive and
  prevents most regressions
  
  [Where problems could occur]
  TBD
+ 
+ [Other info]
+ Originally reported at 
https://www.reddit.com/r/Kubuntu/comments/n9vg9e/cant_make_backports_repo_for_hirsute_to_work/

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1928088

Title:
  Empty pockets don't end up in the cache - E: The value 'hirsute-
  backports' is invalid for APT::Default-Release as such a release is
  not available in the sources

Status in apt package in Ubuntu:
  New

Bug description:
  [Impact]
  If a pocket is empty, apt will not download the empty packages files, causing 
the repository to not be added to the cache. This in turn breaks commands such 
as using apt install -t hirsute-backports as apt is unable to find the release 
in the archive.

  [Test plan]
  Assuming an empty pocket is around, use apt install -t - to 
see it's in the cache.

  An appropriate test case for empty stuff should be added to the test
  suite as well (FIXME before SRUing), the test suite is extensive and
  prevents most regressions

  [Where problems could occur]
  TBD

  [Other info]
  Originally reported at 
https://www.reddit.com/r/Kubuntu/comments/n9vg9e/cant_make_backports_repo_for_hirsute_to_work/

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1928088/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926437] Re: [SRU] Backport zstd support, fix bug in python-apt

2021-05-11 Thread Julian Andres Klode
The xenial SRU should be good to go now, the snapcraft failures are
unrelated and also reproduced in vorlon's retry of snapcraft itself.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926437

Title:
  [SRU] Backport zstd support, fix bug in python-apt

Status in apt package in Ubuntu:
  Fix Released
Status in python-apt package in Ubuntu:
  New
Status in apt source package in Xenial:
  Fix Released
Status in python-apt source package in Xenial:
  Fix Committed
Status in python-apt source package in Bionic:
  Fix Committed
Status in python-apt source package in Focal:
  Fix Committed
Status in python-apt source package in Groovy:
  Fix Committed
Status in python-apt source package in Hirsute:
  Fix Committed
Status in python-apt source package in Impish:
  New

Bug description:
  [Impact]
  APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.

  python-apt in all releases needs to be adjusted to pass the compressor
  names instead of programs to ExtractTar, as otherwise, if
  /usr/bin/zstd does not exist, it will pass "false" as the compressor,
  and then ExtractTar fails (same for other compressors).

  On hirsute, the python-apt upload also includes some new  
annotations for build-dependencies, as it's a straight upload of the 2.2
  stable series.

  [Test plan]
  The apt regression tests cover the compression methods extensively, including 
even stored compressed lists in /var/lib/apt and fetching Packages files 
compressed with such methods, for all compression methods.

  The final test is to make apt_inst.DebFile("glibc-doc-
  reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall() from
  1923845 work in python3-apt.

  For hirsute, the build dependency changes will verify themselves
  during the build ;)

  [Where problems could occur]

  APT:

  The changes to APT are adding the zstd compression backend to FileFd
  and then linking it into the configuration. The former is adding new
  code, triggered only on zstd files, and should not be able to cause
  regressions.

  The latter is a bunch of small lines here and there, and it caused
  issues during backporting in the test suite because it expected
  compressors in a specific order. However, chances are that nothing
  else has as stringent requirements as the test suite, so I don't
  believe that is a problem for existing users.

  Bugs can of course be in the code dealing with zstd, and that might
  open you up to security vulnerabilities in the zstd code if you pass
  APT zstd compressed files.

  A regression can be if clients used apt-helper cat-file to cat a zstd-
  compressed file ending in .zst - they'll now get uncompressed content,
  rather than compressed.

  python-apt:

  The change just replaces Binary with Name, aligning the code with APT
  itself, so it's so trivial, we don't expect regressions.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926437/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923464] Re: ideally should boot rootfs from a matching hard drive

2021-05-07 Thread Julian Andres Klode
Just to not forget: This should work. We pass the ESP/boot partition to
kernel, initramfs then looks at it, and if it finds UUID on same drive,
it will take that one. If not it will take UUID from any drive, thus
handling RAID or encryption automagically.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to systemd in Ubuntu.
https://bugs.launchpad.net/bugs/1923464

Title:
  ideally should boot rootfs from a matching hard drive

Status in cloud-images:
  New
Status in grub2 package in Ubuntu:
  New
Status in initramfs-tools package in Ubuntu:
  New
Status in systemd package in Ubuntu:
  New
Status in u-boot package in Ubuntu:
  New
Status in u-boot-menu package in Ubuntu:
  New

Bug description:
  Ideally we should strive to boot rootfs from a matching hard drive.

  I.e. if we are booting rootfs by UUID, we should try to find the one
  that came from the same drive as where ESP (UEFI) came from, or u-boot
  spl / u-boot got loaded from (loader1/loader2).

  Such that for example, when booted from external usb stick, rootfs
  from there is mounted.

  Or when booted from internal drive whilst a dd backup is attached over
  usb, rootfs is loaded from the internal drive not from the usb
  attached backup.

  This would need:

  * u-boot to export the drive it loaded extlinux.conf / bootscript
  from, and pass it on kernel command line

  * grub to export the device UUID it got loaded from (from the
  BootServices EFI table) and pass it on the kernel command line or via
  runtime EFI variable

  * sdboot already does that I believe, but not sure if initramfs-tools
  consumes the sdboot provided information

  * initramfs-tools to consume above and sort the discovered devices
  based on that, when deciding what to mount as rootfs

To manage notifications about this bug go to:
https://bugs.launchpad.net/cloud-images/+bug/1923464/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1927078] Re: Don't allow useradd to use fully numeric names

2021-05-04 Thread Julian Andres Klode
Maybe it should be a warning in the SRUs as opposed to a failure, but I
don't have a strong opinion. I'm a bit scared of breaking scripts. But
maybe that's a good thing.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to shadow in Ubuntu.
https://bugs.launchpad.net/bugs/1927078

Title:
  Don't allow useradd to use fully numeric names

Status in shadow package in Ubuntu:
  New
Status in shadow source package in Focal:
  New
Status in shadow source package in Groovy:
  New
Status in shadow source package in Hirsute:
  New
Status in shadow source package in Impish:
  New

Bug description:
  [Description]

  Fully numeric names support in Ubuntu is inconsistent in Focal onwards
  because systemd does not like them[1] but are still allowed by default
  by useradd, leaving the session behavior in hands of the running
  applications. Two examples:

  1. After creating a user named "0", the user can log in via ssh or
  console but loginctl won't create a session for it:

  root@focal:/home/ubuntu# useradd -m 0
  root@focal:/home/ubuntu# id 0
  uid=1005(0) gid=1005(0) groups=1005(0)

  ..

  0@192.168.122.6's password:
  Welcome to Ubuntu 20.04.2 LTS (GNU/Linux 5.8.0-48-generic x86_64)

  Last login: Thu Apr  8 16:17:06 2021 from 192.168.122.1
  $ loginctl
  No sessions.
  $ w
   16:20:09 up 4 min,  1 user,  load average: 0.03, 0.14, 0.08
  USER TTY  FROM LOGIN@   IDLE   JCPU   PCPU WHAT
  0pts/0192.168.122.116:170.00s  0.00s  0.00s w  

  And pam-systemd shows the following message:

  Apr 08 16:17:06 focal sshd[1584]: pam_unix(sshd:session): session opened for 
user 0 by (uid=0)
  Apr 08 16:17:06 focal sshd[1584]: pam_systemd(sshd:session): pam-systemd 
initializing
  Apr 08 16:17:06 focal sshd[1584]: pam_systemd(sshd:session): Failed to get 
user record: Invalid argument

  
  2. With that same username, every successful authentication in gdm will loop 
back to gdm again instead of starting gnome, making the user unable to login.

  
  Making useradd fail (unless --badnames is set) when a fully numeric name is 
used will make the default OS behavior consistent.

  
  [Other info]

  - Upstream does not support fully numeric usernames
  - useradd has a --badnames parameter that would still allow the use of these 
type of names

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/shadow/+bug/1927078/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926964] Re: missing libapt-pkg4.12 dependency

2021-05-03 Thread Julian Andres Klode
** Changed in: python-apt (Ubuntu)
 Assignee: (unassigned) => Steve Langasek (vorlon)

** Package changed: python-apt (Ubuntu) => apt (Ubuntu)

** Changed in: apt (Ubuntu)
   Status: New => Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to python-apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926964

Title:
  missing libapt-pkg4.12 dependency

Status in apt package in Ubuntu:
  Triaged

Bug description:
  It seems python-apt from ubuntu precise ESM made it into the old
  precise-security repo. Unfortunately, it depends on libapt-pkg4.12 (>=
  0.8.16~exp12ubuntu10.28), which is not available.

  The following packages have unmet dependencies:
   python-apt : Depends: libapt-pkg4.12 (>= 0.8.16~exp12ubuntu10.28) but 
0.8.16~exp12ubuntu10.27 is to be installed

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926964/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926437] Re: [SRU] Backport zstd support, fix bug in python-apt

2021-04-29 Thread Julian Andres Klode
FWIW, I upgraded python-apt to the version in xenial-proposed

Unpacking python3-apt (1.1.0~beta1ubuntu0.16.04.12) over
(1.1.0~beta1ubuntu0.16.04.11) ...

And verified that the script

import apt_inst, apt_pkg
apt_pkg.init()
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()


worked.

So verification-done-xenial is valid for this one too :D

apport fails weirdly, not have many other autopkgtest results back yet.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926437

Title:
  [SRU] Backport zstd support, fix bug in python-apt

Status in apt package in Ubuntu:
  Invalid
Status in python-apt package in Ubuntu:
  New
Status in apt source package in Xenial:
  Fix Released
Status in python-apt source package in Xenial:
  New
Status in python-apt source package in Bionic:
  New
Status in python-apt source package in Focal:
  New
Status in python-apt source package in Groovy:
  New
Status in python-apt source package in Hirsute:
  New
Status in python-apt source package in Impish:
  New

Bug description:
  [Impact]
  APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.

  python-apt in all releases needs to be adjusted to pass the compressor
  names instead of programs to ExtractTar, as otherwise, if
  /usr/bin/zstd does not exist, it will pass "false" as the compressor,
  and then ExtractTar fails (same for other compressors).

  On hirsute, the python-apt upload also includes some new  
annotations for build-dependencies, as it's a straight upload of the 2.2
  stable series.

  [Test plan]
  The apt regression tests cover the compression methods extensively, including 
even stored compressed lists in /var/lib/apt and fetching Packages files 
compressed with such methods, for all compression methods.

  The final test is to make apt_inst.DebFile("glibc-doc-
  reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall() from
  1923845 work in python3-apt.

  For hirsute, the build dependency changes will verify themselves
  during the build ;)

  [Where problems could occur]

  APT:

  The changes to APT are adding the zstd compression backend to FileFd
  and then linking it into the configuration. The former is adding new
  code, triggered only on zstd files, and should not be able to cause
  regressions.

  The latter is a bunch of small lines here and there, and it caused
  issues during backporting in the test suite because it expected
  compressors in a specific order. However, chances are that nothing
  else has as stringent requirements as the test suite, so I don't
  believe that is a problem for existing users.

  Bugs can of course be in the code dealing with zstd, and that might
  open you up to security vulnerabilities in the zstd code if you pass
  APT zstd compressed files.

  A regression can be if clients used apt-helper cat-file to cat a zstd-
  compressed file ending in .zst - they'll now get uncompressed content,
  rather than compressed.

  python-apt:

  The change just replaces Binary with Name, aligning the code with APT
  itself, so it's so trivial, we don't expect regressions.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926437/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926437] Re: [SRU] Backport zstd support, fix bug in python-apt

2021-04-29 Thread Julian Andres Klode
autopkgtests for apt have passed. I've build a patched version of
python-apt and verified that it worked with the apt version, so the apt
side is verified now :D

** Tags removed: verification-needed verification-needed-xenial
** Tags added: verification-done verification-done-xenial

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926437

Title:
  [SRU] Backport zstd support, fix bug in python-apt

Status in apt package in Ubuntu:
  Invalid
Status in python-apt package in Ubuntu:
  New
Status in apt source package in Xenial:
  Fix Committed
Status in python-apt source package in Xenial:
  New
Status in python-apt source package in Bionic:
  New
Status in python-apt source package in Focal:
  New
Status in python-apt source package in Groovy:
  New
Status in python-apt source package in Hirsute:
  New
Status in python-apt source package in Impish:
  New

Bug description:
  [Impact]
  APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.

  python-apt in all releases needs to be adjusted to pass the compressor
  names instead of programs to ExtractTar, as otherwise, if
  /usr/bin/zstd does not exist, it will pass "false" as the compressor,
  and then ExtractTar fails (same for other compressors).

  On hirsute, the python-apt upload also includes some new  
annotations for build-dependencies, as it's a straight upload of the 2.2
  stable series.

  [Test plan]
  The apt regression tests cover the compression methods extensively, including 
even stored compressed lists in /var/lib/apt and fetching Packages files 
compressed with such methods, for all compression methods.

  The final test is to make apt_inst.DebFile("glibc-doc-
  reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall() from
  1923845 work in python3-apt.

  For hirsute, the build dependency changes will verify themselves
  during the build ;)

  [Where problems could occur]

  APT:

  The changes to APT are adding the zstd compression backend to FileFd
  and then linking it into the configuration. The former is adding new
  code, triggered only on zstd files, and should not be able to cause
  regressions.

  The latter is a bunch of small lines here and there, and it caused
  issues during backporting in the test suite because it expected
  compressors in a specific order. However, chances are that nothing
  else has as stringent requirements as the test suite, so I don't
  believe that is a problem for existing users.

  Bugs can of course be in the code dealing with zstd, and that might
  open you up to security vulnerabilities in the zstd code if you pass
  APT zstd compressed files.

  A regression can be if clients used apt-helper cat-file to cat a zstd-
  compressed file ending in .zst - they'll now get uncompressed content,
  rather than compressed.

  python-apt:

  The change just replaces Binary with Name, aligning the code with APT
  itself, so it's so trivial, we don't expect regressions.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926437/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926150] Re: [SRU] Backport JSON hooks 0.2

2021-04-29 Thread Julian Andres Klode
** Tags removed: verification-needed verification-needed-xenial
** Tags added: verification-done verification-done-xenial

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926150

Title:
  [SRU] Backport JSON hooks 0.2

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Xenial:
  Fix Committed
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Hirsute:
  New
Status in apt source package in Impish:
  Fix Committed

Bug description:
  [Impact]
  We want to be able to use JSON hooks 0.2 in xenial and later releases to 
display messages to users during dist-upgrade.

  For xenial, we backport the JSON hook support in its entirety, whereas
  for later releases, we only need to cherry-pick bugfixes and the v0.2
  improvements.

  [Test plan]
  We have included unit and integration tests for the new JSON code to test the 
writer and prevent regressions in future changes. Regressions in other parts of 
the code should be caught by the test suite.

  
  [Where problems could occur]
  There may be bugs in the JSON code (apt-private/private-json-hooks.*) and in 
the integration thereof. The JSON hook code is identical to 2.3.2, whereas the 
integration code needed slight adjustments.

  The integration changes are limited to adding the hooks, and (for
  xenial, already in later releases) collecting all unknown package
  names. We believe the regression potential of these changes to be
  fairly limited.

  However, errors in JSON hooks are fatal, so once JSON hooks are
  running, they can cause apt to fail, which has to be kept in mind.

  Existing JSON hooks might not expect the new notifications, and break,
  but we are only aware of the snapd hook and it ignores any unknown
  notifications, so that should not pose a problem.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1925745] Re: Hirsute: Cannot install ubuntu-desktop due to unmet dependencies

2021-04-29 Thread Julian Andres Klode
Or we phase _everything_. Also makes code faster

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Impish)
   Importance: Undecided
   Status: Confirmed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1925745

Title:
  Hirsute: Cannot install ubuntu-desktop due to unmet dependencies

Status in apt package in Ubuntu:
  Confirmed
Status in apt source package in Hirsute:
  New
Status in apt source package in Impish:
  Confirmed

Bug description:
  I'm using the ubuntu-21.04-live-server-amd64 ISO to install my hirsute
  desktop machines so I can automate the installation.

  After installation, I'm trying to install the desktop via "apt install
  ubuntu-desktop".

  This fails with:
  "Some packages could not be installed. This may mean that you have
  requested an impossible situation or if you are using the unstable
  distribution that some required packages have not yet been created
  or been moved out of Incoming.
  The following information may help to resolve the situation:

  The following packages have unmet dependencies:
   ubuntu-release-upgrader-gtk : Depends: ubuntu-release-upgrader-core (= 
1:21.04.11) but 1:21.04.10 is to be installed
 Depends: python3-distupgrade (= 1:21.04.11) 
but 1:21.04.10 is to be installed
  E: Unable to correct problems, you have held broken packages."

  "apt policy ubuntu-release-upgrader-core" shows:
  "ubuntu-release-upgrader-core:
Installed: 1:21.04.10
Candidate: 1:21.04.10
Version table:
   1:21.04.11 1 (phased 20%)
  500 http://at.archive.ubuntu.com/ubuntu hirsute-updates/main amd64 
Packages
   *** 1:21.04.10 500
  500 http://at.archive.ubuntu.com/ubuntu hirsute/main amd64 Packages
  100 /var/lib/dpkg/status"

  
  I could not find any information what "phased 20%" means. It's obviously a 
major problem not being able to install the desktop on a fresh installation.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1925745/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1925745] Re: Hirsute: Cannot install ubuntu-desktop due to unmet dependencies

2021-04-29 Thread Julian Andres Klode
OK, so what's going on is that ubuntu-release-upgrader-gtk is not
installed, and apt only updates phasing to upgrades, but not new
installs. Which fails in this case. Presumably it needs to apply phasing
to all packages in a source package if one of them is installed, but
that might be slow :(

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1925745

Title:
  Hirsute: Cannot install ubuntu-desktop due to unmet dependencies

Status in apt package in Ubuntu:
  Confirmed

Bug description:
  I'm using the ubuntu-21.04-live-server-amd64 ISO to install my hirsute
  desktop machines so I can automate the installation.

  After installation, I'm trying to install the desktop via "apt install
  ubuntu-desktop".

  This fails with:
  "Some packages could not be installed. This may mean that you have
  requested an impossible situation or if you are using the unstable
  distribution that some required packages have not yet been created
  or been moved out of Incoming.
  The following information may help to resolve the situation:

  The following packages have unmet dependencies:
   ubuntu-release-upgrader-gtk : Depends: ubuntu-release-upgrader-core (= 
1:21.04.11) but 1:21.04.10 is to be installed
 Depends: python3-distupgrade (= 1:21.04.11) 
but 1:21.04.10 is to be installed
  E: Unable to correct problems, you have held broken packages."

  "apt policy ubuntu-release-upgrader-core" shows:
  "ubuntu-release-upgrader-core:
Installed: 1:21.04.10
Candidate: 1:21.04.10
Version table:
   1:21.04.11 1 (phased 20%)
  500 http://at.archive.ubuntu.com/ubuntu hirsute-updates/main amd64 
Packages
   *** 1:21.04.10 500
  500 http://at.archive.ubuntu.com/ubuntu hirsute/main amd64 Packages
  100 /var/lib/dpkg/status"

  
  I could not find any information what "phased 20%" means. It's obviously a 
major problem not being able to install the desktop on a fresh installation.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1925745/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926437] Re: [SRU] Backport zstd support, fix bug in python-apt

2021-04-29 Thread Julian Andres Klode
** Description changed:

  [Impact]
  APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.
  
  python-apt in all releases needs to be adjusted to pass the compressor
  names instead of programs to ExtractTar, as otherwise, if /usr/bin/zstd
  does not exist, it will pass "false" as the compressor, and then
  ExtractTar fails (same for other compressors).
  
+ On hirsute, the python-apt upload also includes some new  
annotations for build-dependencies, as it's a straight upload of the 2.2
+ stable series.
+ 
  [Test plan]
  The apt regression tests cover the compression methods extensively, including 
even stored compressed lists in /var/lib/apt and fetching Packages files 
compressed with such methods, for all compression methods.
  
  The final test is to make apt_inst.DebFile("glibc-doc-
  reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall() from
  1923845 work in python3-apt.
+ 
+ For hirsute, the build dependency changes will verify themselves during
+ the build ;)
  
  [Where problems could occur]
  
  APT:
  
  The changes to APT are adding the zstd compression backend to FileFd and
  then linking it into the configuration. The former is adding new code,
  triggered only on zstd files, and should not be able to cause
  regressions.
  
  The latter is a bunch of small lines here and there, and it caused
  issues during backporting in the test suite because it expected
  compressors in a specific order. However, chances are that nothing else
  has as stringent requirements as the test suite, so I don't believe that
  is a problem for existing users.
  
  Bugs can of course be in the code dealing with zstd, and that might open
  you up to security vulnerabilities in the zstd code if you pass APT zstd
  compressed files.
  
  A regression can be if clients used apt-helper cat-file to cat a zstd-
  compressed file ending in .zst - they'll now get uncompressed content,
  rather than compressed.
  
- 
  python-apt:
  
  The change just replaces Binary with Name, aligning the code with APT
  itself, so it's so trivial, we don't expect regressions.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926437

Title:
  [SRU] Backport zstd support, fix bug in python-apt

Status in apt package in Ubuntu:
  Invalid
Status in python-apt package in Ubuntu:
  New
Status in apt source package in Xenial:
  Fix Committed
Status in python-apt source package in Xenial:
  New
Status in python-apt source package in Bionic:
  New
Status in python-apt source package in Focal:
  New
Status in python-apt source package in Groovy:
  New
Status in python-apt source package in Hirsute:
  New
Status in python-apt source package in Impish:
  New

Bug description:
  [Impact]
  APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.

  python-apt in all releases needs to be adjusted to pass the compressor
  names instead of programs to ExtractTar, as otherwise, if
  /usr/bin/zstd does not exist, it will pass "false" as the compressor,
  and then ExtractTar fails (same for other compressors).

  On hirsute, the python-apt upload also includes some new  
annotations for build-dependencies, as it's a straight upload of the 2.2
  stable series.

  [Test plan]
  The apt regression tests cover the compression methods extensively, including 
even stored compressed lists in /var/lib/apt and fetching Packages files 
compressed with such methods, for all compression methods.

  The final test is to make apt_inst.DebFile("glibc-doc-
  reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall() from
  1923845 work in python3-apt.

  For hirsute, the build dependency changes will verify themselves
  during the build ;)

  [Where problems could occur]

  APT:

  The changes to APT are adding the zstd compression backend to FileFd
  and then linking it into the configuration. The former is adding new
  code, triggered only on zstd files, and should not be able to cause
  regressions.

  The latter is a bunch of small lines here and there, and it caused
  issues during backporting in the test suite because it expected
  compressors in a specific order. However, chances are that nothing
  else has as stringent requirements as the test suite, so I don't
  believe that is a problem for existing users.

  Bugs can of course be in the code dealing with zstd, and that might
  open you up to security vulnerabilities in the zstd code if you pass
  APT zstd compressed files.

  A regression can be if clients used apt-helper cat-file to cat a zstd-
  compressed file ending in .zst - they'll now get uncompressed content,
  rather than compressed.

  python-apt:

  The change just replaces Binary with Name, aligning the code with APT
  itself, so it's so trivial, we don't expect regressions.

To manage notifications about this bug go to:

[Touch-packages] [Bug 1921626] Re: size mismatch error if request of unknown size is larger than others

2021-04-28 Thread Julian Andres Klode
Yes, see the tasks.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  size mismatch error if request of unknown size is larger than others

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Groovy:
  New
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]

  Downloads fail if:

  - there is at least one package each with Size and no size on a mirror
  - a package without a Size field is larger than a package with a Size field 
that's currently in the pipeline

  Also, this was silent, we need to add an error so people fix repos.
  For hirsute, this is a warning; for hirsute+1 it's an error so
  people's CI fails on them and doesn't "succeed with warnings"

  [Test plan]

  We have included a test case in the apt integration tests, which
  downloads three packages a, b, c where b is largest and has no Size
  field. With 2.2.2, it fails; with 2.2.3 it succceeds.

  We have also added a test case that a warning is shown.

  [Where problems could occur]
  Problems can only occur if you try to download packages without a Size field, 
as that is the only place code changes (adding code guarded with if ... Size 
... == 0; 0 being unknown size).

  [Other changes]
  2.2.3 includes the same change as 2.2.2ubuntu1

  [Original bug report]

  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match.

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926437] [NEW] [SRU] Backport zstd support, fix bug in python-apt

2021-04-28 Thread Julian Andres Klode
Public bug reported:

[Impact]
APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.

python-apt in all releases needs to be adjusted to pass the compressor
names instead of programs to ExtractTar, as otherwise, if /usr/bin/zstd
does not exist, it will pass "false" as the compressor, and then
ExtractTar fails (same for other compressors).

[Test plan]
The apt regression tests cover the compression methods extensively, including 
even stored compressed lists in /var/lib/apt and fetching Packages files 
compressed with such methods, for all compression methods.

The final test is to make apt_inst.DebFile("glibc-doc-
reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall() from
1923845 work in python3-apt.

[Where problems could occur]

APT:

The changes to APT are adding the zstd compression backend to FileFd and
then linking it into the configuration. The former is adding new code,
triggered only on zstd files, and should not be able to cause
regressions.

The latter is a bunch of small lines here and there, and it caused
issues during backporting in the test suite because it expected
compressors in a specific order. However, chances are that nothing else
has as stringent requirements as the test suite, so I don't believe that
is a problem for existing users.

Bugs can of course be in the code dealing with zstd, and that might open
you up to security vulnerabilities in the zstd code if you pass APT zstd
compressed files.

A regression can be if clients used apt-helper cat-file to cat a zstd-
compressed file ending in .zst - they'll now get uncompressed content,
rather than compressed.


python-apt:

The change just replaces Binary with Name, aligning the code with APT
itself, so it's so trivial, we don't expect regressions.

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: Invalid

** Affects: python-apt (Ubuntu)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Xenial)
 Importance: Undecided
 Status: New

** Affects: python-apt (Ubuntu Xenial)
 Importance: Undecided
 Status: New

** Affects: python-apt (Ubuntu Bionic)
 Importance: Undecided
 Status: New

** Affects: python-apt (Ubuntu Focal)
 Importance: Undecided
 Status: New

** Affects: python-apt (Ubuntu Groovy)
 Importance: Undecided
 Status: New

** Affects: python-apt (Ubuntu Hirsute)
 Importance: Undecided
 Status: New

** Affects: python-apt (Ubuntu Impish)
 Importance: Undecided
 Status: New

** Also affects: python-apt (Ubuntu)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: python-apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Impish)
   Importance: Undecided
   Status: New

** Also affects: python-apt (Ubuntu Impish)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Xenial)
   Importance: Undecided
   Status: New

** Also affects: python-apt (Ubuntu Xenial)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

** Also affects: python-apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Groovy)
   Importance: Undecided
   Status: New

** Also affects: python-apt (Ubuntu Groovy)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** Also affects: python-apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** No longer affects: apt (Ubuntu Bionic)

** No longer affects: apt (Ubuntu Focal)

** No longer affects: apt (Ubuntu Groovy)

** No longer affects: apt (Ubuntu Hirsute)

** No longer affects: apt (Ubuntu Impish)

** Changed in: apt (Ubuntu)
   Status: New => Invalid

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926437

Title:
  [SRU] Backport zstd support, fix bug in python-apt

Status in apt package in Ubuntu:
  Invalid
Status in python-apt package in Ubuntu:
  New
Status in apt source package in Xenial:
  New
Status in python-apt source package in Xenial:
  New
Status in python-apt source package in Bionic:
  New
Status in python-apt source package in Focal:
  New
Status in python-apt source package in Groovy:
  New
Status in python-apt source package in Hirsute:
  New
Status in python-apt source package in Impish:
  New

Bug description:
  [Impact]
  APT in xenial needs to learn about zstd support to enable Launchpad to work 
with zstd packages.

  python-apt in all releases needs to be adjusted to pass the compressor
  names instead of programs to ExtractTar, as otherwise, if
  /usr/bin/zstd does not exist, it will pass "false" as the compressor,
  and then ExtractTar 

[Touch-packages] [Bug 1923845] Re: Please compress packages with zstd by default

2021-04-28 Thread Julian Andres Klode
Split out APT / python-apt into bug 1923845

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923845

Title:
  Please compress packages with zstd by default

Status in appstream-glib package in Ubuntu:
  New
Status in apt package in Ubuntu:
  New
Status in aptly package in Ubuntu:
  New
Status in boinc package in Ubuntu:
  New
Status in busybox package in Ubuntu:
  New
Status in cdebootstrap package in Ubuntu:
  New
Status in cdist package in Ubuntu:
  New
Status in debdelta package in Ubuntu:
  New
Status in debian-el package in Ubuntu:
  New
Status in debootstrap package in Ubuntu:
  In Progress
Status in debsig-verify package in Ubuntu:
  New
Status in debsigs package in Ubuntu:
  New
Status in diffoscope package in Ubuntu:
  New
Status in dpkg package in Ubuntu:
  New
Status in dpkg-sig package in Ubuntu:
  New
Status in file package in Ubuntu:
  New
Status in libsolv package in Ubuntu:
  New
Status in lintian package in Ubuntu:
  New
Status in lutris package in Ubuntu:
  New
Status in obs-build package in Ubuntu:
  New
Status in osc package in Ubuntu:
  New
Status in radare2 package in Ubuntu:
  New
Status in reprepro package in Ubuntu:
  New
Status in vim-scripts package in Ubuntu:
  New
Status in zeroinstall-injector package in Ubuntu:
  New

Bug description:
  https://people.canonical.com/~rbalint/zstd-debs/ contains a .deb built
  on Hirsute having both data and control members of the .deb being
  compressed with zstd. It can be handy for testing various tools.

  [dpkg]
  Decompression support in dpkg landed first in Bionic and is being SRUd to 
Xenial in LP: #1764220 enable Launchpad's Xenial systems to process the 
zstd-compressed binary packages.
  From dpkg's perspective the upgrade path is cleared.

  The original plan was compressing only the internal data.tar .deb
  member, but dpkg uses uniform compression by default since dpkg 1.19.0
  thus I'm collecting all the changes to support control.tar.zst, too,
  in this bug.

  Reviewed packages from:
  https://codesearch.debian.net/search?q=data.tar.xz=1=1
  https://codesearch.debian.net/search?q=control.tar.xz=1=1

  appstream-glib  - needs fix: libappstream-builder/asb-package-deb.c
  aptly   - needs fix: deb/deb.go
  boinc   - needs fix: debian/fetch_example_applications.sh
  busybox - needs fix: archival/dpkg_deb.c archival/dpkg.c
  cdebootstrap- needs fix: src/package.c
  cdist   - may need fix, can use dpkg-deb: 
cdist/preos/debootstrap/files/devuan-debootstrap/functions
  debdelta- needs fix: debdelta debpatch.sh
  debian-el   - needs fix: deb-view.el
  debian-handbook - needs fix, maybe later, for Debian
  debootstrap - needs fix, 
https://salsa.debian.org/installer-team/debootstrap/-/merge_requests/54
  debsigs - needs fix, debsigs
  debsig-verify   - needs fix, src/debsig-verify.c
  diffoscope  - needs fix, diffoscope/comparators/deb.py
  dpkg- needs fix, change default
  dpkg-sig- needs fix, dpkg-sig
  dpmb- needs fix, maybe later, for Debian
  elfutils- may need fix, uses dpkg-deb if it is available, does not 
handle .gz either
  file- needs fix, magic/Magdir/archive
  libsolv - needs fix, ext/repo_deb.c
  lintian - needs fix malformed-deb-archive
  lutris  - needs fix, lutris/util/extract.py
  obs-build   - needs fix Build/Deb.pm
  osc - needs fix osc/util/debquery.py control.tar.zst only
  python-apt  - needs fix 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()
  radare2 - needs fix
  reprepro- needs fix, debfile.c
  vim-scripts - needs fix debPlugin/autoload/deb.vim
  winetricks  - needs fix when Debian switches src/winetricks
  zeroinstall-injector - needs fix src/zeroinstall/archive.ml

  acr - skip, does not _have to_ be fixed, just creates packages, 
see dist/deb_hand.mak
  alien   - skip, uses dpkg-deb to extract .deb
  ansible - not affected, just test data in dbdata.tar.xz
  anthy   - not affected, just changelog entry
  apt - seems fixed already
  ceph- not affected in Ubuntu's version
  circlator   - not affected, just test data
  cowdancer   - not affected, just documentation
  eccodes - skip, just orig-data.tar.xz
  eckit   - skip, just ...orig-data.tar.xz
  firefox - skip, profdata.tar.xz
  firefox-esr - skip, profdata.tar.xz
  galculator  - skip, just changelog
  grads   - skip, ...orig-data.tar.xz
  gvmd- skip, just creates xz compressed .deb
  insighttoolkit4 - skip, ...orig-data.tar.xz
  jdeb- skip, just creates compressed .deb packages
  jmol- skip, just local data archive
  jq  - skip, extracting .deb in Tracis CI only
 

[Touch-packages] [Bug 1923845] Re: Please compress packages with zstd by default

2021-04-26 Thread Julian Andres Klode
Aside from the code being wrong - it's missing an apt_pkg.init_config(),
it also does not work with the config being setup; it needs to be
whitelisted.

So we definitely need to SRU APT. However, we are already preparing
1.2.35 at the moment for ESM enablement, I don't know if we want to
stuff both things together.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923845

Title:
  Please compress packages with zstd by default

Status in appstream-glib package in Ubuntu:
  New
Status in apt package in Ubuntu:
  New
Status in aptly package in Ubuntu:
  New
Status in boinc package in Ubuntu:
  New
Status in busybox package in Ubuntu:
  New
Status in cdebootstrap package in Ubuntu:
  New
Status in cdist package in Ubuntu:
  New
Status in debdelta package in Ubuntu:
  New
Status in debian-el package in Ubuntu:
  New
Status in debootstrap package in Ubuntu:
  In Progress
Status in debsig-verify package in Ubuntu:
  New
Status in debsigs package in Ubuntu:
  New
Status in diffoscope package in Ubuntu:
  New
Status in dpkg package in Ubuntu:
  New
Status in dpkg-sig package in Ubuntu:
  New
Status in file package in Ubuntu:
  New
Status in libsolv package in Ubuntu:
  New
Status in lintian package in Ubuntu:
  New
Status in lutris package in Ubuntu:
  New
Status in obs-build package in Ubuntu:
  New
Status in osc package in Ubuntu:
  New
Status in radare2 package in Ubuntu:
  New
Status in reprepro package in Ubuntu:
  New
Status in vim-scripts package in Ubuntu:
  New
Status in zeroinstall-injector package in Ubuntu:
  New

Bug description:
  https://people.canonical.com/~rbalint/zstd-debs/ contains a .deb built
  on Hirsute having both data and control members of the .deb being
  compressed with zstd. It can be handy for testing various tools.

  [dpkg]
  Decompression support in dpkg landed first in Bionic and is being SRUd to 
Xenial in LP: #1764220 enable Launchpad's Xenial systems to process the 
zstd-compressed binary packages.
  From dpkg's perspective the upgrade path is cleared.

  The original plan was compressing only the internal data.tar .deb
  member, but dpkg uses uniform compression by default since dpkg 1.19.0
  thus I'm collecting all the changes to support control.tar.zst, too,
  in this bug.

  Reviewed packages from:
  https://codesearch.debian.net/search?q=data.tar.xz=1=1
  https://codesearch.debian.net/search?q=control.tar.xz=1=1

  appstream-glib  - needs fix: libappstream-builder/asb-package-deb.c
  aptly   - needs fix: deb/deb.go
  boinc   - needs fix: debian/fetch_example_applications.sh
  busybox - needs fix: archival/dpkg_deb.c archival/dpkg.c
  cdebootstrap- needs fix: src/package.c
  cdist   - may need fix, can use dpkg-deb: 
cdist/preos/debootstrap/files/devuan-debootstrap/functions
  debdelta- needs fix: debdelta debpatch.sh
  debian-el   - needs fix: deb-view.el
  debian-handbook - needs fix, maybe later, for Debian
  debootstrap - needs fix, 
https://salsa.debian.org/installer-team/debootstrap/-/merge_requests/54
  debsigs - needs fix, debsigs
  debsig-verify   - needs fix, src/debsig-verify.c
  diffoscope  - needs fix, diffoscope/comparators/deb.py
  dpkg- needs fix, change default
  dpkg-sig- needs fix, dpkg-sig
  dpmb- needs fix, maybe later, for Debian
  elfutils- may need fix, uses dpkg-deb if it is available, does not 
handle .gz either
  file- needs fix, magic/Magdir/archive
  libsolv - needs fix, ext/repo_deb.c
  lintian - needs fix malformed-deb-archive
  lutris  - needs fix, lutris/util/extract.py
  obs-build   - needs fix Build/Deb.pm
  osc - needs fix osc/util/debquery.py control.tar.zst only
  python-apt  - needs fix 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()
  radare2 - needs fix
  reprepro- needs fix, debfile.c
  vim-scripts - needs fix debPlugin/autoload/deb.vim
  winetricks  - needs fix when Debian switches src/winetricks
  zeroinstall-injector - needs fix src/zeroinstall/archive.ml

  acr - skip, does not _have to_ be fixed, just creates packages, 
see dist/deb_hand.mak
  alien   - skip, uses dpkg-deb to extract .deb
  ansible - not affected, just test data in dbdata.tar.xz
  anthy   - not affected, just changelog entry
  apt - seems fixed already
  ceph- not affected in Ubuntu's version
  circlator   - not affected, just test data
  cowdancer   - not affected, just documentation
  eccodes - skip, just orig-data.tar.xz
  eckit   - skip, just ...orig-data.tar.xz
  firefox - skip, profdata.tar.xz
  firefox-esr - skip, profdata.tar.xz
  galculator  - skip, just changelog
  grads   - skip, ...orig-data.tar.xz
  

[Touch-packages] [Bug 1876035] Re: Automatically retry downloads on failures

2021-04-26 Thread Julian Andres Klode
The basic "turn the flag on" is landing in 2.3.2, I'll open other bug
reports for the backoff and server switching things I suppose.

** Changed in: apt (Ubuntu)
   Status: Triaged => Fix Committed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1876035

Title:
  Automatically retry downloads on failures

Status in apt package in Ubuntu:
  Fix Committed

Bug description:
  Hi,

  We operate archive.ubuntu.com as well as mirrors of it in various
  places. We're getting reports of various CI/CD jobs failing. Sometimes
  due to routing issues, issues with overloaded servers/VMs hosting
  archive, etc.

  Any chance we can make '-o Acquire::Retries=3' the default for 'apt-
  get update/install/dist-upgrade'? Preferably with a randomised skew
  between retries.

  Also, any chance of having apt-get retry with different IPs/hosts
  returned from the DNS lookup? (e.g. for archive.ubuntu.com -
  91.189.88.142, 91.189.88.152, ...).

  Thanks.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1876035/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926150] [NEW] [SRU] Backport JSON hooks 0.2

2021-04-26 Thread Julian Andres Klode
Public bug reported:

[Impact]
We want to be able to use JSON hooks 0.2 in xenial and later releases to 
display messages to users during dist-upgrade.

For xenial, we backport the JSON hook support in its entirety, whereas
for later releases, we only need to cherry-pick bugfixes and the v0.2
improvements.

[Test plan]
We have included unit and integration tests for the new JSON code to test the 
writer and prevent regressions in future changes. Regressions in other parts of 
the code should be caught by the test suite.


[Where problems could occur]
There may be bugs in the JSON code (apt-private/private-json-hooks.*) and in 
the integration thereof. The JSON hook code is identical to 2.3.2, whereas the 
integration code needed slight adjustments.

The integration changes are limited to adding the hooks, and (for
xenial, already in later releases) collecting all unknown package names.
We believe the regression potential of these changes to be fairly
limited.

However, errors in JSON hooks are fatal, so once JSON hooks are running,
they can cause apt to fail, which has to be kept in mind.

Existing JSON hooks might not expect the new notifications, and break,
but we are only aware of the snapd hook and it ignores any unknown
notifications, so that should not pose a problem.

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: Fix Committed

** Affects: apt (Ubuntu Xenial)
 Importance: Undecided
 Status: Triaged

** Affects: apt (Ubuntu Bionic)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Focal)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Hirsute)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Impish)
 Importance: Undecided
 Status: Fix Committed

** Also affects: apt (Ubuntu Impish)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Xenial)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

** Changed in: apt (Ubuntu Impish)
   Status: New => Fix Committed

** Changed in: apt (Ubuntu Xenial)
   Status: New => Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926150

Title:
  [SRU] Backport JSON hooks 0.2

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Xenial:
  Triaged
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Hirsute:
  New
Status in apt source package in Impish:
  Fix Committed

Bug description:
  [Impact]
  We want to be able to use JSON hooks 0.2 in xenial and later releases to 
display messages to users during dist-upgrade.

  For xenial, we backport the JSON hook support in its entirety, whereas
  for later releases, we only need to cherry-pick bugfixes and the v0.2
  improvements.

  [Test plan]
  We have included unit and integration tests for the new JSON code to test the 
writer and prevent regressions in future changes. Regressions in other parts of 
the code should be caught by the test suite.

  
  [Where problems could occur]
  There may be bugs in the JSON code (apt-private/private-json-hooks.*) and in 
the integration thereof. The JSON hook code is identical to 2.3.2, whereas the 
integration code needed slight adjustments.

  The integration changes are limited to adding the hooks, and (for
  xenial, already in later releases) collecting all unknown package
  names. We believe the regression potential of these changes to be
  fairly limited.

  However, errors in JSON hooks are fatal, so once JSON hooks are
  running, they can cause apt to fail, which has to be kept in mind.

  Existing JSON hooks might not expect the new notifications, and break,
  but we are only aware of the snapd hook and it ignores any unknown
  notifications, so that should not pose a problem.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1926150] Re: [SRU] Backport JSON hooks 0.2

2021-04-26 Thread Julian Andres Klode
These changes are in apt 2.3.2 which is queued for impish.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1926150

Title:
  [SRU] Backport JSON hooks 0.2

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Xenial:
  Triaged
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Hirsute:
  New
Status in apt source package in Impish:
  Fix Committed

Bug description:
  [Impact]
  We want to be able to use JSON hooks 0.2 in xenial and later releases to 
display messages to users during dist-upgrade.

  For xenial, we backport the JSON hook support in its entirety, whereas
  for later releases, we only need to cherry-pick bugfixes and the v0.2
  improvements.

  [Test plan]
  We have included unit and integration tests for the new JSON code to test the 
writer and prevent regressions in future changes. Regressions in other parts of 
the code should be caught by the test suite.

  
  [Where problems could occur]
  There may be bugs in the JSON code (apt-private/private-json-hooks.*) and in 
the integration thereof. The JSON hook code is identical to 2.3.2, whereas the 
integration code needed slight adjustments.

  The integration changes are limited to adding the hooks, and (for
  xenial, already in later releases) collecting all unknown package
  names. We believe the regression potential of these changes to be
  fairly limited.

  However, errors in JSON hooks are fatal, so once JSON hooks are
  running, they can cause apt to fail, which has to be kept in mind.

  Existing JSON hooks might not expect the new notifications, and break,
  but we are only aware of the snapd hook and it ignores any unknown
  notifications, so that should not pose a problem.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1926150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1925833] Re: apt-transport-https reparses URLs, breaking CloudFront signing

2021-04-23 Thread Julian Andres Klode
This has been fixed in 2.1.15 and newer versions, so starting with
hirsute. As the change is too invasive, it has not been backported to
older releases.

** Changed in: apt (Ubuntu)
   Status: New => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1925833

Title:
  apt-transport-https reparses URLs, breaking CloudFront signing

Status in apt package in Ubuntu:
  Fix Released

Bug description:
  CloudFront URL signing requires that the the whole URL matches the signature, 
as compared to S3 URLs which prune the querystring before validating the 
signature. The URL re-parsing which I'm pretty sure was introduced in resolving 
#1651923 decodes some HTML entities which were encoded before signing. That 
invalidates the signature and results in permission denied errors. In my case, 
the equal and semicolon chars from the return-content-disposition header are 
being decoded, but it applies to any entity which is "optionally" encoded.
 
  Personally, I think that the URL returned in a Location: redirect header 
should be handled as-is, and if some piece of garbage http server is generating 
redirects which contain spaces or are otherwise invalid, the problem lies with 
the web server generating invalid redirects and not in apt failing to follow 
broken URLs.  I say that knowing full well that the http spec says those things 
should be identical, so CloudFront also falls into the general "garbage" 
category by breaking the spec by requiring a specific format for identical 
characters.

  Either way, things are what they are.  Rather than unnecessarily fully
  reencoding the URL, I'd suggest that apt-transport-https (and
  presumably the http transport as well, but I don't know) should at
  most just replace the spaces with plusses or %20s to keep the original
  bug resolved without breaking other stuff. :)

  PS: I tried to report this against the apt-transport-https package,
  but the bug tool says that doesn't exist in Ubuntu.  That's weird,
  considering it's definitely a package
  (https://packages.ubuntu.com/groovy/apt-transport-https), so I assume
  the apt maintainer will know how to triage. Sorry.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1925833/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1925833] Re: apt-transport-https reparses URLs, breaking CloudFront signing

2021-04-23 Thread Julian Andres Klode
WRT your other comment: apt-transport-https is a binary package,
launchpad bugs are against source packages. Also apt-transport-https is
an empty transitional package, there is no separate https method
anymore.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1925833

Title:
  apt-transport-https reparses URLs, breaking CloudFront signing

Status in apt package in Ubuntu:
  Fix Released

Bug description:
  CloudFront URL signing requires that the the whole URL matches the signature, 
as compared to S3 URLs which prune the querystring before validating the 
signature. The URL re-parsing which I'm pretty sure was introduced in resolving 
#1651923 decodes some HTML entities which were encoded before signing. That 
invalidates the signature and results in permission denied errors. In my case, 
the equal and semicolon chars from the return-content-disposition header are 
being decoded, but it applies to any entity which is "optionally" encoded.
 
  Personally, I think that the URL returned in a Location: redirect header 
should be handled as-is, and if some piece of garbage http server is generating 
redirects which contain spaces or are otherwise invalid, the problem lies with 
the web server generating invalid redirects and not in apt failing to follow 
broken URLs.  I say that knowing full well that the http spec says those things 
should be identical, so CloudFront also falls into the general "garbage" 
category by breaking the spec by requiring a specific format for identical 
characters.

  Either way, things are what they are.  Rather than unnecessarily fully
  reencoding the URL, I'd suggest that apt-transport-https (and
  presumably the http transport as well, but I don't know) should at
  most just replace the spaces with plusses or %20s to keep the original
  bug resolved without breaking other stuff. :)

  PS: I tried to report this against the apt-transport-https package,
  but the bug tool says that doesn't exist in Ubuntu.  That's weird,
  considering it's definitely a package
  (https://packages.ubuntu.com/groovy/apt-transport-https), so I assume
  the apt maintainer will know how to triage. Sorry.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1925833/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923845] Re: Please compress packages with zstd by default

2021-04-21 Thread Julian Andres Klode
I have not tested hirsute in a clean container; but touch /usr/bin/zstd
makes it work, so the compressor lookup is broken.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923845

Title:
  Please compress packages with zstd by default

Status in appstream-glib package in Ubuntu:
  New
Status in apt package in Ubuntu:
  New
Status in aptly package in Ubuntu:
  New
Status in boinc package in Ubuntu:
  New
Status in busybox package in Ubuntu:
  New
Status in cdebootstrap package in Ubuntu:
  New
Status in cdist package in Ubuntu:
  New
Status in debdelta package in Ubuntu:
  New
Status in debian-el package in Ubuntu:
  New
Status in debootstrap package in Ubuntu:
  In Progress
Status in debsig-verify package in Ubuntu:
  New
Status in debsigs package in Ubuntu:
  New
Status in diffoscope package in Ubuntu:
  New
Status in dpkg package in Ubuntu:
  New
Status in dpkg-sig package in Ubuntu:
  New
Status in file package in Ubuntu:
  New
Status in libsolv package in Ubuntu:
  New
Status in lintian package in Ubuntu:
  New
Status in lutris package in Ubuntu:
  New
Status in obs-build package in Ubuntu:
  New
Status in osc package in Ubuntu:
  New
Status in radare2 package in Ubuntu:
  New
Status in reprepro package in Ubuntu:
  New
Status in vim-scripts package in Ubuntu:
  New
Status in zeroinstall-injector package in Ubuntu:
  New

Bug description:
  https://people.canonical.com/~rbalint/zstd-debs/ contains a .deb built
  on Hirsute having both data and control members of the .deb being
  compressed with zstd. It can be handy for testing various tools.

  [dpkg]
  Decompression support in dpkg landed first in Bionic and is being SRUd to 
Xenial in LP: #1764220 enable Launchpad's Xenial systems to process the 
zstd-compressed binary packages.
  From dpkg's perspective the upgrade path is cleared.

  The original plan was compressing only the internal data.tar .deb
  member, but dpkg uses uniform compression by default since dpkg 1.19.0
  thus I'm collecting all the changes to support control.tar.zst, too,
  in this bug.

  Reviewed packages from:
  https://codesearch.debian.net/search?q=data.tar.xz=1=1
  https://codesearch.debian.net/search?q=control.tar.xz=1=1

  appstream-glib  - needs fix: libappstream-builder/asb-package-deb.c
  aptly   - needs fix: deb/deb.go
  boinc   - needs fix: debian/fetch_example_applications.sh
  busybox - needs fix: archival/dpkg_deb.c archival/dpkg.c
  cdebootstrap- needs fix: src/package.c
  cdist   - may need fix, can use dpkg-deb: 
cdist/preos/debootstrap/files/devuan-debootstrap/functions
  debdelta- needs fix: debdelta debpatch.sh
  debian-el   - needs fix: deb-view.el
  debian-handbook - needs fix, maybe later, for Debian
  debootstrap - needs fix, 
https://salsa.debian.org/installer-team/debootstrap/-/merge_requests/54
  debsigs - needs fix, debsigs
  debsig-verify   - needs fix, src/debsig-verify.c
  diffoscope  - needs fix, diffoscope/comparators/deb.py
  dpkg- needs fix, change default
  dpkg-sig- needs fix, dpkg-sig
  dpmb- needs fix, maybe later, for Debian
  elfutils- may need fix, uses dpkg-deb if it is available, does not 
handle .gz either
  file- needs fix, magic/Magdir/archive
  libsolv - needs fix, ext/repo_deb.c
  lintian - needs fix malformed-deb-archive
  lutris  - needs fix, lutris/util/extract.py
  obs-build   - needs fix Build/Deb.pm
  osc - needs fix osc/util/debquery.py control.tar.zst only
  python-apt  - needs fix 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()
  radare2 - needs fix
  reprepro- needs fix, debfile.c
  vim-scripts - needs fix debPlugin/autoload/deb.vim
  winetricks  - needs fix when Debian switches src/winetricks
  zeroinstall-injector - needs fix src/zeroinstall/archive.ml

  acr - skip, does not _have to_ be fixed, just creates packages, 
see dist/deb_hand.mak
  alien   - skip, uses dpkg-deb to extract .deb
  ansible - not affected, just test data in dbdata.tar.xz
  anthy   - not affected, just changelog entry
  apt - seems fixed already
  ceph- not affected in Ubuntu's version
  circlator   - not affected, just test data
  cowdancer   - not affected, just documentation
  eccodes - skip, just orig-data.tar.xz
  eckit   - skip, just ...orig-data.tar.xz
  firefox - skip, profdata.tar.xz
  firefox-esr - skip, profdata.tar.xz
  galculator  - skip, just changelog
  grads   - skip, ...orig-data.tar.xz
  gvmd- skip, just creates xz compressed .deb
  insighttoolkit4 - skip, ...orig-data.tar.xz
  jdeb- skip, just creates compressed .deb packages
  jmol- skip, just 

[Touch-packages] [Bug 1923845] Re: Please compress packages with zstd by default

2021-04-21 Thread Julian Andres Klode
There seems to be a bug in bionic's apt.

root@bb:/home/jak/Projects# python3 -c 'import apt_inst, apt_pkg; 
apt_pkg.init(); 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2_zstd1_all.deb").control.extractdata("control")'
Traceback (most recent call last):
  File "", line 1, in 
LookupError: There is no member named 'control'
Error in sys.excepthook:
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/apport_python_hook.py", line 72, in 
apport_excepthook
from apport.fileutils import likely_packaged, get_recent_crashes
  File "/usr/lib/python3/dist-packages/apport/__init__.py", line 5, in 
from apport.report import Report
  File "/usr/lib/python3/dist-packages/apport/report.py", line 30, in 
import apport.fileutils
  File "/usr/lib/python3/dist-packages/apport/fileutils.py", line 25, in 

from apport.packaging_impl import impl as packaging
  File "/usr/lib/python3/dist-packages/apport/packaging_impl.py", line 24, in 

import apt
  File "/usr/lib/python3/dist-packages/apt/__init__.py", line 35, in 
apt_pkg.init_system()
apt_pkg.Error: E:Cannot find a configured compressor for 'false'


** No longer affects: python-apt (Ubuntu)

** Also affects: apt (Ubuntu)
   Importance: Undecided
   Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to busybox in Ubuntu.
https://bugs.launchpad.net/bugs/1923845

Title:
  Please compress packages with zstd by default

Status in appstream-glib package in Ubuntu:
  New
Status in apt package in Ubuntu:
  New
Status in aptly package in Ubuntu:
  New
Status in boinc package in Ubuntu:
  New
Status in busybox package in Ubuntu:
  New
Status in cdebootstrap package in Ubuntu:
  New
Status in cdist package in Ubuntu:
  New
Status in debdelta package in Ubuntu:
  New
Status in debian-el package in Ubuntu:
  New
Status in debootstrap package in Ubuntu:
  In Progress
Status in debsig-verify package in Ubuntu:
  New
Status in debsigs package in Ubuntu:
  New
Status in diffoscope package in Ubuntu:
  New
Status in dpkg package in Ubuntu:
  New
Status in dpkg-sig package in Ubuntu:
  New
Status in file package in Ubuntu:
  New
Status in libsolv package in Ubuntu:
  New
Status in lintian package in Ubuntu:
  New
Status in lutris package in Ubuntu:
  New
Status in obs-build package in Ubuntu:
  New
Status in osc package in Ubuntu:
  New
Status in radare2 package in Ubuntu:
  New
Status in reprepro package in Ubuntu:
  New
Status in vim-scripts package in Ubuntu:
  New
Status in zeroinstall-injector package in Ubuntu:
  New

Bug description:
  https://people.canonical.com/~rbalint/zstd-debs/ contains a .deb built
  on Hirsute having both data and control members of the .deb being
  compressed with zstd. It can be handy for testing various tools.

  [dpkg]
  Decompression support in dpkg landed first in Bionic and is being SRUd to 
Xenial in LP: #1764220 enable Launchpad's Xenial systems to process the 
zstd-compressed binary packages.
  From dpkg's perspective the upgrade path is cleared.

  The original plan was compressing only the internal data.tar .deb
  member, but dpkg uses uniform compression by default since dpkg 1.19.0
  thus I'm collecting all the changes to support control.tar.zst, too,
  in this bug.

  Reviewed packages from:
  https://codesearch.debian.net/search?q=data.tar.xz=1=1
  https://codesearch.debian.net/search?q=control.tar.xz=1=1

  appstream-glib  - needs fix: libappstream-builder/asb-package-deb.c
  aptly   - needs fix: deb/deb.go
  boinc   - needs fix: debian/fetch_example_applications.sh
  busybox - needs fix: archival/dpkg_deb.c archival/dpkg.c
  cdebootstrap- needs fix: src/package.c
  cdist   - may need fix, can use dpkg-deb: 
cdist/preos/debootstrap/files/devuan-debootstrap/functions
  debdelta- needs fix: debdelta debpatch.sh
  debian-el   - needs fix: deb-view.el
  debian-handbook - needs fix, maybe later, for Debian
  debootstrap - needs fix, 
https://salsa.debian.org/installer-team/debootstrap/-/merge_requests/54
  debsigs - needs fix, debsigs
  debsig-verify   - needs fix, src/debsig-verify.c
  diffoscope  - needs fix, diffoscope/comparators/deb.py
  dpkg- needs fix, change default
  dpkg-sig- needs fix, dpkg-sig
  dpmb- needs fix, maybe later, for Debian
  elfutils- may need fix, uses dpkg-deb if it is available, does not 
handle .gz either
  file- needs fix, magic/Magdir/archive
  libsolv - needs fix, ext/repo_deb.c
  lintian - needs fix malformed-deb-archive
  lutris  - needs fix, lutris/util/extract.py
  obs-build   - needs fix Build/Deb.pm
  osc - needs fix osc/util/debquery.py control.tar.zst only
  python-apt  - needs fix 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()
  radare2 - needs fix
  reprepro- 

[Touch-packages] [Bug 1923845] Re: Please compress packages with zstd by default

2021-04-21 Thread Julian Andres Klode
I don't understand what python-apt does here, it inherits zstd support
from apt.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to busybox in Ubuntu.
https://bugs.launchpad.net/bugs/1923845

Title:
  Please compress packages with zstd by default

Status in appstream-glib package in Ubuntu:
  New
Status in apt package in Ubuntu:
  New
Status in aptly package in Ubuntu:
  New
Status in boinc package in Ubuntu:
  New
Status in busybox package in Ubuntu:
  New
Status in cdebootstrap package in Ubuntu:
  New
Status in cdist package in Ubuntu:
  New
Status in debdelta package in Ubuntu:
  New
Status in debian-el package in Ubuntu:
  New
Status in debootstrap package in Ubuntu:
  In Progress
Status in debsig-verify package in Ubuntu:
  New
Status in debsigs package in Ubuntu:
  New
Status in diffoscope package in Ubuntu:
  New
Status in dpkg package in Ubuntu:
  New
Status in dpkg-sig package in Ubuntu:
  New
Status in file package in Ubuntu:
  New
Status in libsolv package in Ubuntu:
  New
Status in lintian package in Ubuntu:
  New
Status in lutris package in Ubuntu:
  New
Status in obs-build package in Ubuntu:
  New
Status in osc package in Ubuntu:
  New
Status in radare2 package in Ubuntu:
  New
Status in reprepro package in Ubuntu:
  New
Status in vim-scripts package in Ubuntu:
  New
Status in zeroinstall-injector package in Ubuntu:
  New

Bug description:
  https://people.canonical.com/~rbalint/zstd-debs/ contains a .deb built
  on Hirsute having both data and control members of the .deb being
  compressed with zstd. It can be handy for testing various tools.

  [dpkg]
  Decompression support in dpkg landed first in Bionic and is being SRUd to 
Xenial in LP: #1764220 enable Launchpad's Xenial systems to process the 
zstd-compressed binary packages.
  From dpkg's perspective the upgrade path is cleared.

  The original plan was compressing only the internal data.tar .deb
  member, but dpkg uses uniform compression by default since dpkg 1.19.0
  thus I'm collecting all the changes to support control.tar.zst, too,
  in this bug.

  Reviewed packages from:
  https://codesearch.debian.net/search?q=data.tar.xz=1=1
  https://codesearch.debian.net/search?q=control.tar.xz=1=1

  appstream-glib  - needs fix: libappstream-builder/asb-package-deb.c
  aptly   - needs fix: deb/deb.go
  boinc   - needs fix: debian/fetch_example_applications.sh
  busybox - needs fix: archival/dpkg_deb.c archival/dpkg.c
  cdebootstrap- needs fix: src/package.c
  cdist   - may need fix, can use dpkg-deb: 
cdist/preos/debootstrap/files/devuan-debootstrap/functions
  debdelta- needs fix: debdelta debpatch.sh
  debian-el   - needs fix: deb-view.el
  debian-handbook - needs fix, maybe later, for Debian
  debootstrap - needs fix, 
https://salsa.debian.org/installer-team/debootstrap/-/merge_requests/54
  debsigs - needs fix, debsigs
  debsig-verify   - needs fix, src/debsig-verify.c
  diffoscope  - needs fix, diffoscope/comparators/deb.py
  dpkg- needs fix, change default
  dpkg-sig- needs fix, dpkg-sig
  dpmb- needs fix, maybe later, for Debian
  elfutils- may need fix, uses dpkg-deb if it is available, does not 
handle .gz either
  file- needs fix, magic/Magdir/archive
  libsolv - needs fix, ext/repo_deb.c
  lintian - needs fix malformed-deb-archive
  lutris  - needs fix, lutris/util/extract.py
  obs-build   - needs fix Build/Deb.pm
  osc - needs fix osc/util/debquery.py control.tar.zst only
  python-apt  - needs fix 
apt_inst.DebFile("glibc-doc-reference_2.33-0ubuntu2~zstd1_all.deb").control.extractall()
  radare2 - needs fix
  reprepro- needs fix, debfile.c
  vim-scripts - needs fix debPlugin/autoload/deb.vim
  winetricks  - needs fix when Debian switches src/winetricks
  zeroinstall-injector - needs fix src/zeroinstall/archive.ml

  acr - skip, does not _have to_ be fixed, just creates packages, 
see dist/deb_hand.mak
  alien   - skip, uses dpkg-deb to extract .deb
  ansible - not affected, just test data in dbdata.tar.xz
  anthy   - not affected, just changelog entry
  apt - seems fixed already
  ceph- not affected in Ubuntu's version
  circlator   - not affected, just test data
  cowdancer   - not affected, just documentation
  eccodes - skip, just orig-data.tar.xz
  eckit   - skip, just ...orig-data.tar.xz
  firefox - skip, profdata.tar.xz
  firefox-esr - skip, profdata.tar.xz
  galculator  - skip, just changelog
  grads   - skip, ...orig-data.tar.xz
  gvmd- skip, just creates xz compressed .deb
  insighttoolkit4 - skip, ...orig-data.tar.xz
  jdeb- skip, just creates compressed .deb packages
  jmol- skip, just local data archive
  jq  - 

[Touch-packages] [Bug 1923358] Re: Other servers doesn't display server list (Software Updater)

2021-04-20 Thread Julian Andres Klode
** Package changed: update-manager (Ubuntu) => software-properties
(Ubuntu)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to software-properties in
Ubuntu.
https://bugs.launchpad.net/bugs/1923358

Title:
  Other servers doesn't display server list (Software Updater)

Status in software-properties package in Ubuntu:
  Incomplete

Bug description:
  Using Settings and then trying to choose other server doesn't do
  anything. It should display the list of available other servers.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/software-properties/+bug/1923358/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923464] Re: ideally should boot rootfs from a matching hard drive

2021-04-20 Thread Julian Andres Klode
This proposal seems somewhat incompatible with our resilient boot
specification.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to systemd in Ubuntu.
https://bugs.launchpad.net/bugs/1923464

Title:
  ideally should boot rootfs from a matching hard drive

Status in cloud-images:
  New
Status in grub2 package in Ubuntu:
  New
Status in initramfs-tools package in Ubuntu:
  New
Status in systemd package in Ubuntu:
  New
Status in u-boot package in Ubuntu:
  New
Status in u-boot-menu package in Ubuntu:
  New

Bug description:
  Ideally we should strive to boot rootfs from a matching hard drive.

  I.e. if we are booting rootfs by UUID, we should try to find the one
  that came from the same drive as where ESP (UEFI) came from, or u-boot
  spl / u-boot got loaded from (loader1/loader2).

  Such that for example, when booted from external usb stick, rootfs
  from there is mounted.

  Or when booted from internal drive whilst a dd backup is attached over
  usb, rootfs is loaded from the internal drive not from the usb
  attached backup.

  This would need:

  * u-boot to export the drive it loaded extlinux.conf / bootscript
  from, and pass it on kernel command line

  * grub to export the device UUID it got loaded from (from the
  BootServices EFI table) and pass it on the kernel command line or via
  runtime EFI variable

  * sdboot already does that I believe, but not sure if initramfs-tools
  consumes the sdboot provided information

  * initramfs-tools to consume above and sort the discovered devices
  based on that, when deciding what to mount as rootfs

To manage notifications about this bug go to:
https://bugs.launchpad.net/cloud-images/+bug/1923464/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1925025] Re: package packagekit 1.1.13-2ubuntu1.1 failed to install/upgrade: package is in a very bad inconsistent state; you should reinstall it before attempting configuration

2021-04-20 Thread Julian Andres Klode
Thanks for your bug report. Unfortunately, it lacks any actionable
information, because we don't know why packagekit is in a very bad
inconsistent state.

If you figure out the cause for this, please let us know.

Thanks!

** Changed in: packagekit (Ubuntu)
   Status: New => Incomplete

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to packagekit in Ubuntu.
https://bugs.launchpad.net/bugs/1925025

Title:
  package packagekit 1.1.13-2ubuntu1.1 failed to install/upgrade:
  package is in a very bad inconsistent state; you should  reinstall it
  before attempting configuration

Status in packagekit package in Ubuntu:
  Incomplete

Bug description:
  no further information

  ProblemType: Package
  DistroRelease: Ubuntu 20.04
  Package: packagekit 1.1.13-2ubuntu1.1
  ProcVersionSignature: Ubuntu 5.8.0-45.51~20.04.1-generic 5.8.18
  Uname: Linux 5.8.0-45-generic x86_64
  ApportVersion: 2.20.11-0ubuntu27.16
  Architecture: amd64
  CasperMD5CheckResult: skip
  Date: Mon Apr 19 14:17:46 2021
  DuplicateSignature:
   package:packagekit:1.1.13-2ubuntu1.1
   Setting up libcurl4:amd64 (7.68.0-1ubuntu2.5) ...
   dpkg: error processing package packagekit (--configure):
package is in a very bad inconsistent state; you should
  ErrorMessage: package is in a very bad inconsistent state; you should  
reinstall it before attempting configuration
  InstallationDate: Installed on 2020-09-22 (209 days ago)
  InstallationMedia: Ubuntu 20.04.1 LTS "Focal Fossa" - Release amd64 (20200731)
  Python3Details: /usr/bin/python3.8, Python 3.8.5, python3-minimal, 
3.8.2-0ubuntu2
  PythonDetails: /usr/bin/python2.7, Python 2.7.18, python-is-python2, 2.7.17-4
  RelatedPackageVersions:
   dpkg 1.19.7ubuntu3
   apt  2.0.5
  SourcePackage: packagekit
  Title: package packagekit 1.1.13-2ubuntu1.1 failed to install/upgrade: 
package is in a very bad inconsistent state; you should  reinstall it before 
attempting configuration
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/packagekit/+bug/1925025/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923150] Re: unexpected error pop-up after 'apt install nvidia-driver-455'

2021-04-18 Thread Julian Andres Klode
Yes, of course if you don't install the fixed version you still have the
broken one.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923150

Title:
  unexpected error pop-up after 'apt install nvidia-driver-455'

Status in OEM Priority Project:
  Confirmed
Status in apt package in Ubuntu:
  Invalid
Status in nvidia-graphics-drivers-455 package in Ubuntu:
  Invalid

Bug description:
  Ubuntu info:

  $ lsb_release -rd
  Description:Ubuntu 20.04.1 LTS (fossa-spark-cml X52)
  Release 20.04

  Full reproduce step

  1. nvidia-driver-455 455.45.01-0ubuntu0.20.04.1 is installed in the first 
place, without all i386 pkg installed.
  2. run update-manager to do upgrade.
  3. it will show error after download deb and didn't upgrade anything.

  Per test with the command to do the upgrade, it will have an extra
  error message, but it will just upgrade and install recommended i386
  debs.

  The error msgs are:

  E: Could not configure 'libc6:i386'.
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  Short way to get the error:
  # apt-get install libgcc-s1:i386 --dry-run
  Reading package lists... Done
  Building dependency tree   
  Reading state information... Done
  The following packages were automatically installed and are no longer 
required:
apt-clone archdetect-deb dctrl-tools dmraid gir1.2-timezonemap-1.0 
gir1.2-xkl-1.0 kpartx kpartx-boot libdebian-installer4 libdmraid1.0.0.rc16
libtimezonemap-data libtimezonemap1 python3-icu python3-pam rdate
  Use 'apt autoremove' to remove them.
  The following additional packages will be installed:
gcc-10-base:i386 libc6 libc6:i386 libcrypt1:i386 libidn2-0:i386 
libunistring2:i386
  Suggested packages:
glibc-doc glibc-doc:i386 locales:i386
  The following NEW packages will be installed:
gcc-10-base:i386 libc6:i386 libcrypt1:i386 libgcc-s1:i386 libidn2-0:i386 
libunistring2:i386
  The following packages will be upgraded:
libc6
  1 upgraded, 6 newly installed, 0 to remove and 235 not upgraded.
  Inst libc6 [2.31-0ubuntu9.1] (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates 
[amd64])
  Inst libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386]) []
  Conf libc6 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [amd64]) []
  Inst libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Conf gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386])
  Conf libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386])
  Inst libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Inst libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  Conf libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Conf libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  E: Could not configure 'libc6:i386'. 
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  For now, we think the above Error is the root cause of update-manager
  failure.

  ---

  This happens as upgrade package nvidia-driver-455 from
  455.45.01-0ubuntu0.20.04.1 to 460.39-0ubuntu0.20.04.1.

  for more detail, please check the attachment.

To manage notifications about this bug go to:
https://bugs.launchpad.net/oem-priority/+bug/1923150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1924809] Re: apt 2.0.5 in focal causes autopkgtest failures for update-manager

2021-04-16 Thread Julian Andres Klode
Can confirm it's broken:

^['linux-.*'-5\.4\.0-72-generic$|^ 'kfreebsd-.*'-5\.4\.0-72-generic$|^ 
'gnumach-.*'-5\.4\.0-72-generic$|^ '.*-modules'-5\.4\.0-72-generic$|^ 
'.*-kernel'-5\.4\.0-72-generic$|^ 'linux-.*'-5\.4\.0-72-generic$|^ 
'kfreebsd-.*'-5\.4\.0-72-generic$|^ 'gnumach-.*'-5\
.4\.0-72-generic$|^ '.*-modules'-5\.4\.0-72-generic$|^ 
'.*-kernel'-5\.4\.0-72-generic$|^ 'linux-.*'-5\.4\.0-72-generic$|^ 
'kfreebsd-.*'-5\.4\.0-72-generic$|^ 'gnumach-.*'-5\.4\.0-72-generic$|^ 
'.*-modules'-5\.4\.0-72-generic$|^ '.*-kernel']-5\.4\.0-72-generic$
23:03

Seems like the test is broken


APT::VersionedKernelPackages "['linux-.*', 'kfreebsd-.*', 'gnumach-.*', 
'.*-modules', '.*-kernel', 'linux-.*', 'kfreebsd-.*', 'gnumach-.*', 
'.*-modules', '.*-kernel', 'linux-.*', 'kfreebsd-.*', 'gnumach-.*', 
'.*-modules', '.*-kernel']";

This code sequence is wrong:

# override the kernel autoremove regex
real_versioned_kernel_pkgs = apt.apt_pkg.config.value_list(
"APT::VersionedKernelPackages")
apt.apt_pkg.config.set("APT::VersionedKernelPackages", "linux-.*")
self.addCleanup(
lambda: apt.apt_pkg.config.set("APT::VersionedKernelPackages",
   str(real_versioned_kernel_pkgs)))

It reads a list, then sets a value option instead of the list, and then
after exit, overrides the value option with a string representation of
the python list.

** Package changed: apt (Ubuntu) => update-manager (Ubuntu)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1924809

Title:
  apt 2.0.5 in focal causes autopkgtest failures for update-manager

Status in update-manager package in Ubuntu:
  Triaged

Bug description:
  ==
  ERROR: Test that updates from the security node go in
  --
  Traceback (most recent call last):
File "/tmp/autopkgtest.0o0kua/build.B7h/src/tests/test_update_list.py", 
line 30, in setUp
  self.cache = MyCache(apt.progress.base.OpProgress(),
File "/tmp/autopkgtest.0o0kua/build.B7h/src/UpdateManager/Core/MyCache.py", 
line 70, in __init__
  apt.Cache.__init__(self, progress, rootdir)
File "/usr/lib/python3/dist-packages/apt/cache.py", line 170, in __init__
  self.open(progress)
File "/usr/lib/python3/dist-packages/apt/cache.py", line 232, in open
  self._cache = apt_pkg.Cache(progress)
  apt_pkg.Error: W:Unable to read 
/tmp/autopkgtest.0o0kua/build.B7h/src/tests/aptroot-update-list-test/etc/apt/preferences.d/
 - DirectoryExists (2: No such file or directory), E:Regex compilation error - 
Invalid range end

  Full test results can be found here:

  
https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac
  /autopkgtest-focal/focal/amd64/u/update-
  manager/20210416_180839_df210@/log.gz

  Downgrading apt to version 2.0.4 causes the test to pass again.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/1924809/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923150] Re: unexpected error pop-up after 'apt install nvidia-driver-455'

2021-04-16 Thread Julian Andres Klode
Marking the packaging tags as invalid, as the bug is fixed in apt 2.0.4
which is in the 20.04.2 images. This bug was tracked as bug 1871268 back
when it was being fixed.

** Changed in: apt (Ubuntu)
   Status: Incomplete => Invalid

** Changed in: nvidia-graphics-drivers-455 (Ubuntu)
   Status: Incomplete => Invalid

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923150

Title:
  unexpected error pop-up after 'apt install nvidia-driver-455'

Status in OEM Priority Project:
  Confirmed
Status in apt package in Ubuntu:
  Invalid
Status in nvidia-graphics-drivers-455 package in Ubuntu:
  Invalid

Bug description:
  Ubuntu info:

  $ lsb_release -rd
  Description:Ubuntu 20.04.1 LTS (fossa-spark-cml X52)
  Release 20.04

  Full reproduce step

  1. nvidia-driver-455 455.45.01-0ubuntu0.20.04.1 is installed in the first 
place, without all i386 pkg installed.
  2. run update-manager to do upgrade.
  3. it will show error after download deb and didn't upgrade anything.

  Per test with the command to do the upgrade, it will have an extra
  error message, but it will just upgrade and install recommended i386
  debs.

  The error msgs are:

  E: Could not configure 'libc6:i386'.
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  Short way to get the error:
  # apt-get install libgcc-s1:i386 --dry-run
  Reading package lists... Done
  Building dependency tree   
  Reading state information... Done
  The following packages were automatically installed and are no longer 
required:
apt-clone archdetect-deb dctrl-tools dmraid gir1.2-timezonemap-1.0 
gir1.2-xkl-1.0 kpartx kpartx-boot libdebian-installer4 libdmraid1.0.0.rc16
libtimezonemap-data libtimezonemap1 python3-icu python3-pam rdate
  Use 'apt autoremove' to remove them.
  The following additional packages will be installed:
gcc-10-base:i386 libc6 libc6:i386 libcrypt1:i386 libidn2-0:i386 
libunistring2:i386
  Suggested packages:
glibc-doc glibc-doc:i386 locales:i386
  The following NEW packages will be installed:
gcc-10-base:i386 libc6:i386 libcrypt1:i386 libgcc-s1:i386 libidn2-0:i386 
libunistring2:i386
  The following packages will be upgraded:
libc6
  1 upgraded, 6 newly installed, 0 to remove and 235 not upgraded.
  Inst libc6 [2.31-0ubuntu9.1] (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates 
[amd64])
  Inst libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386]) []
  Conf libc6 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [amd64]) []
  Inst libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Conf gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386])
  Conf libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386])
  Inst libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Inst libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  Conf libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Conf libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  E: Could not configure 'libc6:i386'. 
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  For now, we think the above Error is the root cause of update-manager
  failure.

  ---

  This happens as upgrade package nvidia-driver-455 from
  455.45.01-0ubuntu0.20.04.1 to 460.39-0ubuntu0.20.04.1.

  for more detail, please check the attachment.

To manage notifications about this bug go to:
https://bugs.launchpad.net/oem-priority/+bug/1923150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 310262] Re: APT::Acquire::Retries only applies to archives and source files

2021-04-15 Thread Julian Andres Klode
This was fixed in Version: 1.6~alpha6


** Changed in: apt (Ubuntu)
   Status: Fix Committed => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/310262

Title:
  APT::Acquire::Retries only applies to archives and source files

Status in apt package in Ubuntu:
  Fix Released
Status in apt package in Debian:
  Fix Released

Bug description:
  Binary package hint: apt

  The APT::Acquire::Retries configuration directive for apt only applies
  to source and archive, not the various index files (Packages, Release
  and friends).  Intuitively, you would think that it should, but it
  does not.  As far as I can tell, there is no other equivalent
  directive that would restrict the number of time apt should try to
  fetch an index case before it give up.

  This can trigger a relatively rare problem where apt would try to
  download the same index file again and again, forever until stopped,
  if the download systematically fail abruptly midway.  See LP #291748
  for such an example.  If apt is being run automatically in the
  background, this could end consuming a lot of bandwidth and hammer the
  archive pretty badly before someone notice.

  According to long-standing upstream Debian bug #119544, fixing this
  bug would be a relatively trivial code change, but would break the
  ABI.  As such, I would be curious to know if there is an ABI bump on
  apt roadmap; if yes, would it be possible to nominate this bug for the
  next major release of apt?

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/310262/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923150] Re: unexpected error pop-up after 'apt install nvidia-driver-455'

2021-04-15 Thread Julian Andres Klode
I mean 2.0.4, of course, which is included in the current LTS images
(20.04.2); your image is the previous version (20.04.1)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923150

Title:
  unexpected error pop-up after 'apt install nvidia-driver-455'

Status in OEM Priority Project:
  Confirmed
Status in apt package in Ubuntu:
  Incomplete
Status in nvidia-graphics-drivers-455 package in Ubuntu:
  New

Bug description:
  Ubuntu info:

  $ lsb_release -rd
  Description:Ubuntu 20.04.1 LTS (fossa-spark-cml X52)
  Release 20.04

  Full reproduce step

  1. nvidia-driver-455 455.45.01-0ubuntu0.20.04.1 is installed in the first 
place, without all i386 pkg installed.
  2. run update-manager to do upgrade.
  3. it will show error after download deb and didn't upgrade anything.

  Per test with the command to do the upgrade, it will have an extra
  error message, but it will just upgrade and install recommended i386
  debs.

  The error msgs are:

  E: Could not configure 'libc6:i386'.
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  Short way to get the error:
  # apt-get install libgcc-s1:i386 --dry-run
  Reading package lists... Done
  Building dependency tree   
  Reading state information... Done
  The following packages were automatically installed and are no longer 
required:
apt-clone archdetect-deb dctrl-tools dmraid gir1.2-timezonemap-1.0 
gir1.2-xkl-1.0 kpartx kpartx-boot libdebian-installer4 libdmraid1.0.0.rc16
libtimezonemap-data libtimezonemap1 python3-icu python3-pam rdate
  Use 'apt autoremove' to remove them.
  The following additional packages will be installed:
gcc-10-base:i386 libc6 libc6:i386 libcrypt1:i386 libidn2-0:i386 
libunistring2:i386
  Suggested packages:
glibc-doc glibc-doc:i386 locales:i386
  The following NEW packages will be installed:
gcc-10-base:i386 libc6:i386 libcrypt1:i386 libgcc-s1:i386 libidn2-0:i386 
libunistring2:i386
  The following packages will be upgraded:
libc6
  1 upgraded, 6 newly installed, 0 to remove and 235 not upgraded.
  Inst libc6 [2.31-0ubuntu9.1] (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates 
[amd64])
  Inst libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386]) []
  Conf libc6 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [amd64]) []
  Inst libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Conf gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386])
  Conf libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386])
  Inst libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Inst libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  Conf libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Conf libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  E: Could not configure 'libc6:i386'. 
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  For now, we think the above Error is the root cause of update-manager
  failure.

  ---

  This happens as upgrade package nvidia-driver-455 from
  455.45.01-0ubuntu0.20.04.1 to 460.39-0ubuntu0.20.04.1.

  for more detail, please check the attachment.

To manage notifications about this bug go to:
https://bugs.launchpad.net/oem-priority/+bug/1923150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1923150] Re: unexpected error pop-up after 'apt install nvidia-driver-455'

2021-04-15 Thread Julian Andres Klode
It seems you are not using libapt-pkg6.0 in version 2.1.10ubuntu0.2 or
later? This issue was fixed in that version.

** Changed in: apt (Ubuntu)
   Status: New => Incomplete

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1923150

Title:
  unexpected error pop-up after 'apt install nvidia-driver-455'

Status in OEM Priority Project:
  Confirmed
Status in apt package in Ubuntu:
  Incomplete
Status in nvidia-graphics-drivers-455 package in Ubuntu:
  New

Bug description:
  Ubuntu info:

  $ lsb_release -rd
  Description:Ubuntu 20.04.1 LTS (fossa-spark-cml X52)
  Release 20.04

  Full reproduce step

  1. nvidia-driver-455 455.45.01-0ubuntu0.20.04.1 is installed in the first 
place, without all i386 pkg installed.
  2. run update-manager to do upgrade.
  3. it will show error after download deb and didn't upgrade anything.

  Per test with the command to do the upgrade, it will have an extra
  error message, but it will just upgrade and install recommended i386
  debs.

  The error msgs are:

  E: Could not configure 'libc6:i386'.
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  Short way to get the error:
  # apt-get install libgcc-s1:i386 --dry-run
  Reading package lists... Done
  Building dependency tree   
  Reading state information... Done
  The following packages were automatically installed and are no longer 
required:
apt-clone archdetect-deb dctrl-tools dmraid gir1.2-timezonemap-1.0 
gir1.2-xkl-1.0 kpartx kpartx-boot libdebian-installer4 libdmraid1.0.0.rc16
libtimezonemap-data libtimezonemap1 python3-icu python3-pam rdate
  Use 'apt autoremove' to remove them.
  The following additional packages will be installed:
gcc-10-base:i386 libc6 libc6:i386 libcrypt1:i386 libidn2-0:i386 
libunistring2:i386
  Suggested packages:
glibc-doc glibc-doc:i386 locales:i386
  The following NEW packages will be installed:
gcc-10-base:i386 libc6:i386 libcrypt1:i386 libgcc-s1:i386 libidn2-0:i386 
libunistring2:i386
  The following packages will be upgraded:
libc6
  1 upgraded, 6 newly installed, 0 to remove and 235 not upgraded.
  Inst libc6 [2.31-0ubuntu9.1] (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates 
[amd64])
  Inst libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386]) []
  Conf libc6 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [amd64]) []
  Inst libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Conf gcc-10-base:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386]) []
  Inst libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libgcc-s1:i386 (10.2.0-5ubuntu1~20.04 Ubuntu:20.04/focal-updates, 
Ubuntu:20.04/focal-security [i386])
  Conf libcrypt1:i386 (1:4.4.10-10ubuntu4 Ubuntu:20.04/focal [i386])
  Conf libc6:i386 (2.31-0ubuntu9.2 Ubuntu:20.04/focal-updates [i386])
  Inst libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Inst libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  Conf libunistring2:i386 (0.9.10-2 Ubuntu:20.04/focal [i386])
  Conf libidn2-0:i386 (2.2.0-2 Ubuntu:20.04/focal [i386])
  E: Could not configure 'libc6:i386'. 
  E: Could not perform immediate configuration on 'libgcc-s1:i386'. Please see 
man 5 apt.conf under APT::Immediate-Configure for details. (2)

  For now, we think the above Error is the root cause of update-manager
  failure.

  ---

  This happens as upgrade package nvidia-driver-455 from
  455.45.01-0ubuntu0.20.04.1 to 460.39-0ubuntu0.20.04.1.

  for more detail, please check the attachment.

To manage notifications about this bug go to:
https://bugs.launchpad.net/oem-priority/+bug/1923150/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921626] Re: size mismatch error if request of unknown size is larger than others

2021-04-14 Thread Julian Andres Klode
** Description changed:

+ [Impact]
+ 
+ Downloads fail if:
+ 
+ - there is at least one package each with Size and no size on a mirror
+ - a package without a Size field is larger than a package with a Size field 
that's currently in the pipeline
+ 
+ Also, this was silent, we need to add an error so people fix repos. For
+ hirsute, this is a warning; for hirsute+1 it's an error so people's CI
+ fails on them and doesn't "succeed with warnings"
+ 
+ [Test plan]
+ 
+ We have included a test case in the apt integration tests, which
+ downloads three packages a, b, c where b is largest and has no Size
+ field. With 2.2.2, it fails; with 2.2.3 it succceeds.
+ 
+ We have also added a test case that a warning is shown.
+ 
+ [Where problems could occur]
+ Problems can only occur if you try to download packages without a Size field, 
as that is the only place code changes (adding code guarded with if ... Size 
... == 0; 0 being unknown size).
+ 
+ [Original bug report]
+ 
  1) Ubuntu 18.04.5 LTS
  
  2) apt 1.6.12ubuntu0.2
  
  3) What you expected to happen
  
  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.
  
  4) What happened instead
  
  "apt install ..." fails during the download phase with "File has
  unexpected size "
  
- 
  5) What I've established trying to debug the issue:
  
  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data
  
  More details about the issue:
  https://projects.theforeman.org/issues/32178
  
  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
- It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 
+ It seem that apt is trying to match a request to do wrong response, and size 
doesn't match.
  
  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464 bytes).
  That size is correct for one of the package being downloaded, but
  somehow apt is trying to match to multiple other packages.

** Description changed:

  [Impact]
  
  Downloads fail if:
  
  - there is at least one package each with Size and no size on a mirror
  - a package without a Size field is larger than a package with a Size field 
that's currently in the pipeline
  
  Also, this was silent, we need to add an error so people fix repos. For
  hirsute, this is a warning; for hirsute+1 it's an error so people's CI
  fails on them and doesn't "succeed with warnings"
  
  [Test plan]
  
  We have included a test case in the apt integration tests, which
  downloads three packages a, b, c where b is largest and has no Size
  field. With 2.2.2, it fails; with 2.2.3 it succceeds.
  
  We have also added a test case that a warning is shown.
  
  [Where problems could occur]
  Problems can only occur if you try to download packages without a Size field, 
as that is the only place code changes (adding code guarded with if ... Size 
... == 0; 0 being unknown size).
+ 
+ [Other changes]
+ 2.2.3 includes the same change as 2.2.2ubuntu1
  
  [Original bug report]
  
  1) Ubuntu 18.04.5 LTS
  
  2) apt 1.6.12ubuntu0.2
  
  3) What you expected to happen
  
  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.
  
  4) What happened instead
  
  "apt install ..." fails during the download phase with "File has
  unexpected size "
  
  5) What I've established trying to debug the issue:
  
  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data
  
  More details about the issue:
  https://projects.theforeman.org/issues/32178
  
  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match.
  
  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464 bytes).
  That size is correct for one of the package being downloaded, but
  somehow apt is trying to match to multiple other packages.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.

[Touch-packages] [Bug 1921626] Re: size mismatch error if request of unknown size is larger than others

2021-04-13 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Hirsute)
   Status: Triaged => Fix Committed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  size mismatch error if request of unknown size is larger than others

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Groovy:
  New
Status in apt source package in Hirsute:
  Fix Committed

Bug description:
  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  
  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1854772] Re: /var/lib/dpkg/status has faulty entries

2021-04-12 Thread Julian Andres Klode
The status file seems fine, apt is happy with it, this is either a bug
in apt-listchanges or a specific FS corruption, but haven't seen that
before.

** Package changed: apt (Ubuntu) => apt-listchanges (Ubuntu)

** Changed in: apt-listchanges (Ubuntu)
   Status: New => Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1854772

Title:
  /var/lib/dpkg/status has faulty entries

Status in apt-listchanges package in Ubuntu:
  Triaged
Status in dpkg package in Ubuntu:
  New

Bug description:
  Es wurden 1.109 kB in 0 s geholt (3.865 kB/s).
  Traceback (most recent call last):
File "/usr/share/apt-listchanges/DebianFiles.py", line 117, in readfile
  self.stanzas += [ControlStanza(x) for x in f.read().split('\n\n') if x]
File "/usr/share/apt-listchanges/DebianFiles.py", line 117, in 
  self.stanzas += [ControlStanza(x) for x in f.read().split('\n\n') if x]
File "/usr/share/apt-listchanges/DebianFiles.py", line 70, in __init__
  field, value = line.split(':', 1)
  ValueError: not enough values to unpack (expected 2, got 1)

  The above exception was the direct cause of the following exception:

  Traceback (most recent call last):
File "/usr/bin/apt-listchanges", line 280, in 
  main(config)
File "/usr/bin/apt-listchanges", line 83, in main
  status.readfile('/var/lib/dpkg/status')
File "/usr/share/apt-listchanges/DebianFiles.py", line 120, in readfile
  {'what': file, 'errmsg': str(ex)}) from ex
  RuntimeError: Fehler beim Verarbeiten von »/var/lib/dpkg/status«: not enough 
values to unpack (expected 2, got 1)
  (Lese Datenbank ... 376829 Dateien und Verzeichnisse sind derzeit 
installiert.)

  ProblemType: Bug
  DistroRelease: Ubuntu 19.10
  Package: apt 1.9.4
  ProcVersionSignature: Ubuntu 5.3.0-23.25-generic 5.3.7
  Uname: Linux 5.3.0-23-generic x86_64
  ApportVersion: 2.20.11-0ubuntu8.2
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Mon Dec  2 16:00:56 2019
  InstallationDate: Installed on 2019-09-09 (83 days ago)
  InstallationMedia: Xubuntu 19.04 "Disco Dingo" - Release amd64 (20190416)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt-listchanges/+bug/1854772/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918930] Re: Unexpected file size of one package interrupts update process for all packages and leaves system vulnerable

2021-04-12 Thread Julian Andres Klode
** Changed in: apt (Ubuntu)
   Status: New => Triaged

** Changed in: apt (Ubuntu)
   Importance: Undecided => Low

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918930

Title:
  Unexpected file size of one package interrupts update process for all
  packages and leaves system vulnerable

Status in apt package in Ubuntu:
  Triaged

Bug description:
  An unexpected file size error of *one* package interrupts the whole
  update process for *all* packages and this can leave the system in a
  vulnerable state - this is not a constructed situation, but very real
  right now, look at the following console output - sublime has some
  problems with its package size, but then important ssh updates are not
  executed. Bad.

  The following packages will be upgraded:
brave-browser git git-man libpython2.7-minimal libpython2.7-stdlib 
linux-firmware openssh-client openssh-server openssh-sftp-server python2.7 
python2.7-minimal python3-pil sublime-merge
  13 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
  Need to get 4.548 kB/199 MB of archives.
  After this operation, 1.744 kB of additional disk space will be used.
  Do you want to continue? [Y/n]
  Get:1 https://download.sublimetext.com apt/stable/ sublime-merge 2049 [4.548 
kB]
  Err:1 https://download.sublimetext.com apt/stable/ sublime-merge 2049
File has unexpected size (4542548 != 4548032). Mirror sync in progress? 
[IP: 104.236.0.104 443]
Hashes of expected file:
 - 
SHA512:f65ce3ca80ff0877da48826a0151036cd8e0bdf28b03d225a03f202262ca1278accdac8e7eb46a22904203750ccf06e3abe496a44f7a4b0c3363076501f72369
 - SHA256:e71fcf37e9d934a60b5112a7b79c819f03f55d331371ec0e9b02378c6234478c
 - SHA1:7fe54a9f7ea5383dbdfc0aae39310e2902c6d7f5 [weak]
 - MD5Sum:fd78a3b986bd7da8b2ebd1f659f5938c [weak]
 - Filesize:4548032 [weak]
  E: Failed to fetch 
https://download.sublimetext.com/files/sublime-merge_build-2049_amd64.deb  File 
has unexpected size (4542548 != 4548032). Mirror sync in progress? [IP: 
104.236.0.104 443]
 Hashes of expected file:
  - 
SHA512:f65ce3ca80ff0877da48826a0151036cd8e0bdf28b03d225a03f202262ca1278accdac8e7eb46a22904203750ccf06e3abe496a44f7a4b0c3363076501f72369
  - SHA256:e71fcf37e9d934a60b5112a7b79c819f03f55d331371ec0e9b02378c6234478c
  - SHA1:7fe54a9f7ea5383dbdfc0aae39310e2902c6d7f5 [weak]
  - MD5Sum:fd78a3b986bd7da8b2ebd1f659f5938c [weak]
  - Filesize:4548032 [weak]
  E: Unable to fetch some archives, maybe run apt-get update or try with 
--fix-missing?

  Note: This issue is not about the package size error in a third party
  repo - I do not blame Ubuntu for problems with that. This is about
  breaking the whole process of updating the system because one single
  sub-task fails.

  Why not make the basic tools really robust and reliable?

  BTW - here are s many free pixels on this screen - why not add two
  or three small sentences about text formatting syntax available in
  this extremely primitive text input box? Is there any text formatting
  at all? Why not put just a little bit of love to the user perspective
  and experience? Just two little senteces about formatting would make
  it so much more user friendly to type here. It feels so quick-and-
  dirty, it hurts. Very sad.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918930/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918920] Re: Harden test for no new acquires after transaction abort

2021-04-12 Thread Julian Andres Klode
** Changed in: apt (Ubuntu)
   Status: Fix Committed => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918920

Title:
  Harden test for no new acquires after transaction abort

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Fix Released
Status in apt source package in Focal:
  Fix Released
Status in apt source package in Groovy:
  Fix Released

Bug description:
  [Impact]
  test-pdiff-usage is somewhat flaky, especially on Debian, this makes it less 
flaky. No end user impact as it's a test-only change.

  [Test plan]
  Running autopkgtest, which runs our extensive integration test suite which 
includes the changed test.

  [Where problems could occur]
  No end user regression potential on its own, but might slightly change 
regression potential for future pdiff changes:

  Test approach is slightly different now. It still catches that updates
  fail correctly, but tests more concretely that a transaction was
  aborted rather than that no worker received work (which was not
  guaranteed, the work could be scheduled before it was aborted).

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918920/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918928] Re: APT 2021/03 SRU release scheduling

2021-04-12 Thread Julian Andres Klode
All the SRUs were released the same day, block tags were ignored :-(

** Changed in: apt (Ubuntu)
   Status: In Progress => Invalid

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918928

Title:
  APT 2021/03 SRU release scheduling

Status in apt package in Ubuntu:
  Invalid

Bug description:
  We want to release the APT SRUs from Mar 2021 in a staggered manner,
  such that we have 2-3 days between each release to get more chance to
  discover regressions before rolling out to older releases.

  Hence this bug, which we tag block-proposed-{bionic,focal} and then
  untag once the delay has passed.

  For groovy, we just want the normal aging :-)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918928/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1920601] Re: Frequent test failures caused by networking issues on armhf runners

2021-04-12 Thread Julian Andres Klode
We set Acquire::Retries to 10 on autopkgtest.

** Changed in: apt (Ubuntu)
   Status: New => Invalid

** Changed in: auto-package-testing
   Status: Triaged => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1920601

Title:
  Frequent test failures caused by networking issues on armhf runners

Status in Auto Package Testing:
  Fix Released
Status in apt package in Ubuntu:
  Invalid

Bug description:
  Armhf runners have been known to have networking issues for years.

  A test of Focal's glibc SRU in Bileto shows that those issues did not
  go away. The glibc package itself is not expected to cause regression,
  this was just a run to confirm that.

  The test resuts are shown here:
  
https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_39a8dbb93caf4ec889f8a1b7f69885db/bileto-4017-excuses/2021-03-20_09:25:01/4017_focal_excuses.html

  Out of the 114 failed tests 58 containst the 'Connection timed out'
  string and only test logs from armhf contain that string:

  $ zgrep -l 'Connection timed out' ~/.cache/ubuntu-archive-tools/focal_*.gz 
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_atk1.0_20210318_182825_daefa.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_axe-demultiplexer_20210318_183430_6b008.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_bacula_20210318_190411_e3ad7.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_dialign-t_20210318_204503_b4d94.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_elisa-player_20210318_211129_9537d.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_fyba_20210318_181755_7731e.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gcc-7_20210318_02_4d95c.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gcc-8_20210318_221659_0e69f.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_golang-github-mailru-easyjson_20210318_231917_67888.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gphoto2_20210318_200015_eb767.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gtk+3.0_20210318_235412_db3e5.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_ipmitool_20210319_003215_d03da.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_kbibtex_20210319_004639_0ac1a.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_kcptun_20210318_193309_6dcd9.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libcache-memcached-getparserxs-perl_20210319_024051_33029.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libclass-load-xs-perl_20210319_025846_01419.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libconvert-uulib-perl_20210319_034852_21cd1.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libhat-trie_20210319_060826_fd334.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libhtml-tidy5-perl_20210319_070723_05fa8.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libimager-perl_20210319_063314_b19a2.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libipc-sharelite-perl_20210319_072246_0217e.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_liblexical-var-perl_20210319_070235_4c450.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libmarpa-r2-perl_20210319_073731_f7554.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libmemcached-libmemcached-perl_20210319_011004_dbca1.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libnet-ldns-perl_20210319_081224_49912.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libnet-pcap-perl_20210319_081606_4d772.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libparams-util-perl_20210319_084831_74d13.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libperlio-eol-perl_20210319_085645_97c6c.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libperlio-layers-perl_20210319_085937_8bd70.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libsass_20210319_092518_4db37.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libsdl-perl_20210319_095502_00c8c.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libtext-bibtex-perl_20210319_105027_abf4a.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libtext-hunspell-perl_20210319_103326_4a603.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libunix-syslog-perl_20210319_111403_f4374.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libwx-scintilla-perl_20210319_114818_a6aab.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libzstd_20210319_114737_0c4d9.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_mod-gearman_20210319_132205_d41ca.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_mrbayes_20210319_133633_10adb.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_octave-interval_20210319_151509_8348f.gz
 

[Touch-packages] [Bug 1922786] Re: Apt install of local deb files shows a notification causing confusion

2021-04-12 Thread Julian Andres Klode
** Changed in: apt (Ubuntu)
   Importance: Undecided => Medium

** Changed in: apt (Ubuntu)
   Importance: Medium => Low

** Changed in: apt (Ubuntu)
   Status: New => Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1922786

Title:
  Apt install of local deb files shows a notification causing confusion

Status in apt package in Ubuntu:
  Triaged

Bug description:
  
  Installation for a local deb file shows a notification message that causes a 
lot of confusion whether the package has been successfully installed or not.

  Steps to reproduce
  =
  - Download a package file (xxx.deb)
  - Install the package file from command line "sudo apt install ./xxx.deb"
  - A notification messages shows up at the end of the installation saying:
   "Download is performed unsandboxed as root as file 
'/home/fredli/harddisc/External/Nordic/nRF-Command-Line-Tools_10_12_1_Linux-amd64.deb'
 couldn't be accessed by user '_apt'. - pkgAcquire::Run (13: Permission denied)"

  -> Expected no notification or warnings if installation finished
  successfully.

  It seems that this is only "Notifcation message" BUT it causes a lot
  of confusion whether the package has been installed or not. The
  package seems to be installed correctly from what I understand.

  $ apt --version
  apt 2.2.2 (amd64)

  $ lsb_release -rd
  Description:  Ubuntu Hirsute Hippo (development branch)
  Release:  21.04

  Full log:
  ==
  $ sudo apt install ./nRF-Command-Line-Tools_10_12_1_Linux-amd64.deb 
  Reading package lists... Done
  Building dependency tree... Done
  Reading state information... Done
  Note, selecting 'nrf-command-line-tools' instead of 
'./nRF-Command-Line-Tools_10_12_1_Linux-amd64.deb'
  The following NEW packages will be installed:
nrf-command-line-tools
  0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
  After this operation, 22,6 MB of additional disk space will be used.
  Get:1 
/home/fredli/harddisc/External/Nordic/nRF-Command-Line-Tools_10_12_1_Linux-amd64.deb
 nrf-command-line-tools amd64 10.12.1 [3 406 kB]
  Selecting previously unselected package nrf-command-line-tools.
  (Reading database ... 149444 files and directories currently installed.)
  Preparing to unpack .../nRF-Command-Line-Tools_10_12_1_Linux-amd64.deb ...
  Unpacking nrf-command-line-tools (10.12.1) ...
  Setting up nrf-command-line-tools (10.12.1) ...
  N: Download is performed unsandboxed as root as file 
'/home/fredli/harddisc/External/Nordic/nRF-Command-Line-Tools_10_12_1_Linux-amd64.deb'
 couldn't be accessed by user '_apt'. - pkgAcquire::Run (13: Permission denied)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1922786/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921626] Re: apt install - File has unexpected size - http pipeline

2021-04-08 Thread Julian Andres Klode
I tracked it down further today, accidentally finding a reproducer. In
methods/http.cc we check for Req.DownloadSize > Req.MaximumSize and fail
with an error if that's the case; hence ensuring our download was within
the allowed range for the request.

The issue is that Req.MaximumSize is calculated by
BaseHttpMethod::FindMaximumObjectSizeInQueue as  the maximum of all
maximum sizes in the queue; and requests with unknown sizes are encoded
as 0, so FindMaximumObjectSizeInQueue() will return any other size that
is not 0, despite the 0 meaning "unknown". The fix is to return 0 if any
maximum size is 0.

** Also affects: apt (Ubuntu Groovy)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: Triaged

** Also affects: apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** Summary changed:

- apt install - File has unexpected size - http pipeline
+ size mismatch error if request of unknown size is larger than others

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  size mismatch error if request of unknown size is larger than others

Status in apt package in Ubuntu:
  Triaged
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Groovy:
  New
Status in apt source package in Hirsute:
  Triaged

Bug description:
  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  
  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1922380] Re: remove shim-signed very difficult

2021-04-03 Thread Julian Andres Klode
There is no force option in apt, you can use --force-remove-essential to
avoid enterinh the text but it's obviously dangerous to do so.

As the --force message clearly stated, it comes from dpkg - which is
told to force deinstallation by apt.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1922380

Title:
  remove shim-signed very difficult

Status in apt package in Ubuntu:
  Incomplete

Bug description:
  Hello
  I don't want the shim-signed anymore, but I need an additional password to 
remove it.

  
  This error is for all version of ubuntu.

  I am sure I am entering the correct one because I entered it using a
  copy paste technique.

  Have pity on those who do not know English well !!

  
  apt purge shim-signed
  Lecture des listes de paquets... Fait
  Construction de l'arbre des dépendances   
  Lecture des informations d'état... Fait
  Les paquets suivants ont été installés automatiquement et ne sont plus 
nécessaires :
mokutil shim
  Veuillez utiliser « apt autoremove » pour les supprimer.
  Les paquets suivants seront ENLEVÉS :
shim-signed*
  ATTENTION : Les paquets essentiels suivants vont être enlevés.
  Vous NE devez PAS faire ceci, à moins de savoir exactement ce
  que vous êtes en train de faire.
shim-signed
  0 mis à jour, 0 nouvellement installés, 1 à enlever et 0 non mis à jour.
  Après cette opération, 2 734 ko d'espace disque seront libérés.
  Vous êtes sur le point de faire quelque chose de potentiellement dangereux
  Pour continuer, tapez la phrase « Oui, faites ce que je vous dis ! »
   ?]Oui, faites ce que je vous dis ! 
  Annulation.

  
  root@a:~# export LANG=us

  root@a:~# apt purge shim-signed
  Reading package lists... Done
  Building dependency tree   
  Reading state information... Done
  The following packages were automatically installed and are no longer 
required:
mokutil shim
  Use 'apt autoremove' to remove them.
  The following packages will be REMOVED:
shim-signed*
  WARNING: The following essential packages will be removed.
  This should NOT be done unless you know exactly what you are doing!
shim-signed
  0 upgraded, 0 newly installed, 1 to remove and 171 not upgraded.
  After this operation, 2734 kB disk space will be freed.
  You are about to do something potentially harmful.
  To continue type in the phrase 'Yes, do as I say!'
   ?] Yes, do as I say! 
  perl: warning: Setting locale failed.
  perl: warning: Please check that your locale settings:
LANGUAGE = "fr_FR:fr_CA",
LC_ALL = (unset),
LC_TIME = "fr_FR.UTF-8",
LC_MONETARY = "fr_FR.UTF-8",
LC_ADDRESS = "fr_FR.UTF-8",
LC_TELEPHONE = "fr_FR.UTF-8",
LC_NAME = "fr_FR.UTF-8",
LC_MEASUREMENT = "fr_FR.UTF-8",
LC_IDENTIFICATION = "fr_FR.UTF-8",
LC_NUMERIC = "fr_FR.UTF-8",
LC_PAPER = "fr_FR.UTF-8",
LANG = "us"
  are supported and installed on your system.
  perl: warning: Falling back to the standard locale ("C").
  locale: Cannot set LC_CTYPE to default locale: No such file or directory
  locale: Cannot set LC_MESSAGES to default locale: No such file or directory
  locale: Cannot set LC_ALL to default locale: No such file or directory
  dpkg: warning: overriding problem because --force enabled:
  dpkg: warning: this is a protected package; it should not be removed
  (Reading database ... 211327 files and directories currently installed.)
  Removing shim-signed (1.45+15+1552672080.a4a1fbe-0ubuntu2) ...
  (Reading database ... 211314 files and directories currently installed.)
  Purging configuration files for shim-signed 
(1.45+15+1552672080.a4a1fbe-0ubuntu2) ...
  locale: Cannot set LC_CTYPE to default locale: No such file or directory
  locale: Cannot set LC_MESSAGES to default locale: No such file or directory
  locale: Cannot set LC_ALL to default locale: No such file or directory
  root@a:~#

  ProblemType: Bug
  DistroRelease: Ubuntu 20.10
  Package: evince 3.38.0-1
  ProcVersionSignature: Ubuntu 5.8.0-34.37-generic 5.8.18
  Uname: Linux 5.8.0-34-generic x86_64
  ApportVersion: 2.20.11-0ubuntu50.3
  Architecture: amd64
  CasperMD5CheckResult: pass
  CurrentDesktop: ubuntu:GNOME
  Date: Fri Apr  2 15:07:48 2021
  InstallationDate: Installed on 2020-10-25 (159 days ago)
  InstallationMedia: Ubuntu-Server 20.10 "Groovy Gorilla" - Release amd64 
(20201022)
  KernLog:
   
  SourcePackage: evince
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1922380/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1922380] Re: remove shim-signed very difficult

2021-04-02 Thread Julian Andres Klode
OK it seems you were actually complaining that French did not work for
you. Have you tried entering it manually? Chances are copying messes
things up.

** Changed in: shim-signed (Ubuntu)
   Status: Invalid => Incomplete

** Package changed: shim-signed (Ubuntu) => apt (Ubuntu)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1922380

Title:
  remove shim-signed very difficult

Status in apt package in Ubuntu:
  Incomplete

Bug description:
  Hello
  I don't want the shim-signed anymore, but I need an additional password to 
remove it.

  
  This error is for all version of ubuntu.

  I am sure I am entering the correct one because I entered it using a
  copy paste technique.

  Have pity on those who do not know English well !!

  
  apt purge shim-signed
  Lecture des listes de paquets... Fait
  Construction de l'arbre des dépendances   
  Lecture des informations d'état... Fait
  Les paquets suivants ont été installés automatiquement et ne sont plus 
nécessaires :
mokutil shim
  Veuillez utiliser « apt autoremove » pour les supprimer.
  Les paquets suivants seront ENLEVÉS :
shim-signed*
  ATTENTION : Les paquets essentiels suivants vont être enlevés.
  Vous NE devez PAS faire ceci, à moins de savoir exactement ce
  que vous êtes en train de faire.
shim-signed
  0 mis à jour, 0 nouvellement installés, 1 à enlever et 0 non mis à jour.
  Après cette opération, 2 734 ko d'espace disque seront libérés.
  Vous êtes sur le point de faire quelque chose de potentiellement dangereux
  Pour continuer, tapez la phrase « Oui, faites ce que je vous dis ! »
   ?]Oui, faites ce que je vous dis ! 
  Annulation.

  
  root@a:~# export LANG=us

  root@a:~# apt purge shim-signed
  Reading package lists... Done
  Building dependency tree   
  Reading state information... Done
  The following packages were automatically installed and are no longer 
required:
mokutil shim
  Use 'apt autoremove' to remove them.
  The following packages will be REMOVED:
shim-signed*
  WARNING: The following essential packages will be removed.
  This should NOT be done unless you know exactly what you are doing!
shim-signed
  0 upgraded, 0 newly installed, 1 to remove and 171 not upgraded.
  After this operation, 2734 kB disk space will be freed.
  You are about to do something potentially harmful.
  To continue type in the phrase 'Yes, do as I say!'
   ?] Yes, do as I say! 
  perl: warning: Setting locale failed.
  perl: warning: Please check that your locale settings:
LANGUAGE = "fr_FR:fr_CA",
LC_ALL = (unset),
LC_TIME = "fr_FR.UTF-8",
LC_MONETARY = "fr_FR.UTF-8",
LC_ADDRESS = "fr_FR.UTF-8",
LC_TELEPHONE = "fr_FR.UTF-8",
LC_NAME = "fr_FR.UTF-8",
LC_MEASUREMENT = "fr_FR.UTF-8",
LC_IDENTIFICATION = "fr_FR.UTF-8",
LC_NUMERIC = "fr_FR.UTF-8",
LC_PAPER = "fr_FR.UTF-8",
LANG = "us"
  are supported and installed on your system.
  perl: warning: Falling back to the standard locale ("C").
  locale: Cannot set LC_CTYPE to default locale: No such file or directory
  locale: Cannot set LC_MESSAGES to default locale: No such file or directory
  locale: Cannot set LC_ALL to default locale: No such file or directory
  dpkg: warning: overriding problem because --force enabled:
  dpkg: warning: this is a protected package; it should not be removed
  (Reading database ... 211327 files and directories currently installed.)
  Removing shim-signed (1.45+15+1552672080.a4a1fbe-0ubuntu2) ...
  (Reading database ... 211314 files and directories currently installed.)
  Purging configuration files for shim-signed 
(1.45+15+1552672080.a4a1fbe-0ubuntu2) ...
  locale: Cannot set LC_CTYPE to default locale: No such file or directory
  locale: Cannot set LC_MESSAGES to default locale: No such file or directory
  locale: Cannot set LC_ALL to default locale: No such file or directory
  root@a:~#

  ProblemType: Bug
  DistroRelease: Ubuntu 20.10
  Package: evince 3.38.0-1
  ProcVersionSignature: Ubuntu 5.8.0-34.37-generic 5.8.18
  Uname: Linux 5.8.0-34-generic x86_64
  ApportVersion: 2.20.11-0ubuntu50.3
  Architecture: amd64
  CasperMD5CheckResult: pass
  CurrentDesktop: ubuntu:GNOME
  Date: Fri Apr  2 15:07:48 2021
  InstallationDate: Installed on 2020-10-25 (159 days ago)
  InstallationMedia: Ubuntu-Server 20.10 "Groovy Gorilla" - Release amd64 
(20201022)
  KernLog:
   
  SourcePackage: evince
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1922380/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1806076] Re: unattended-upgrade --help raises UnicodeEncodeError when stdout encoding is ascii

2021-03-30 Thread Julian Andres Klode
I installed 1.6.13 and could verify that the regression wrt --help is
fixed. unattended-upgrades still logged a lot of errors because for some
reason gettext returned utf8 despite being on non-UTF-8 locale, but
that's all from the APT side.

** Tags removed: verification-needed verification-needed-bionic
** Tags added: verification-done verification-done-bionic

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1806076

Title:
  unattended-upgrade --help raises UnicodeEncodeError when stdout
  encoding is ascii

Status in apt package in Ubuntu:
  Fix Released
Status in unattended-upgrades package in Ubuntu:
  Won't Fix
Status in apt source package in Xenial:
  New
Status in unattended-upgrades source package in Xenial:
  Won't Fix
Status in apt source package in Bionic:
  Fix Committed
Status in unattended-upgrades source package in Bionic:
  Won't Fix

Bug description:
  [Impact]

   * unattended-upgrades --help crashes in apt.systemd.daily script when
  locale is not in UTF-8.

  [Test Case]

  In the fixed case no crash is observed.

  rbalint@yogi:~$ lxc launch ubuntu:18.04 bb-lp-1806076
  Creating bb-lp-1806076
  Starting bb-lp-1806076
  rbalint@yogi:~$ lxc shell bb-lp-1806076
  mesg: ttyname failed: No such device
  root@bb-lp-1806076:~# apt install -yqq language-pack-ru-base
  The following package was automatically installed and is no longer required:
    libfreetype6
  Use 'apt autoremove' to remove it.
  The following additional packages will be installed:
    language-pack-ru
  The following NEW packages will be installed:
    language-pack-ru language-pack-ru-base
  0 upgraded, 2 newly installed, 0 to remove and 7 not upgraded.
  Need to get 2310 kB of archives.
  After this operation, 11.8 MB of additional disk space will be used.
  Selecting previously unselected package language-pack-ru-base.
  (Reading database ... 28536 files and directories currently installed.)
  Preparing to unpack .../language-pack-ru-base_1%3a18.04+20180712_all.deb ...
  Unpacking language-pack-ru-base (1:18.04+20180712) ...
  Selecting previously unselected package language-pack-ru.
  Preparing to unpack .../language-pack-ru_1%3a18.04+20180712_all.deb ...
  Unpacking language-pack-ru (1:18.04+20180712) ...
  Setting up language-pack-ru (1:18.04+20180712) ...
  Setting up language-pack-ru-base (1:18.04+20180712) ...
  Generating locales (this might take a while)...
    ru_RU.UTF-8... done
    ru_UA.UTF-8... done
  Generation complete.

  root@bb-lp-1806076:~# echo LANG=ru_RU | tee /etc/default/locale
  LANG=ru_RU
  root@bb-lp-1806076:~#
  /usr/lib/apt/apt.systemd.daily update
  Traceback (most recent call last):
    File "/usr/bin/unattended-upgrade", line 2171, in 
  (options, args) = parser.parse_args()  # type: ignore
    File "/usr/lib/python3.6/optparse.py", line 1387, in parse_args
  stop = self._process_args(largs, rargs, values)
    File "/usr/lib/python3.6/optparse.py", line 1427, in _process_args
  self._process_long_opt(rargs, values)
    File "/usr/lib/python3.6/optparse.py", line 1501, in _process_long_opt
  option.process(opt, value, values, self)
    File "/usr/lib/python3.6/optparse.py", line 785, in process
  self.action, self.dest, opt, value, values, parser)
    File "/usr/lib/python3.6/optparse.py", line 807, in take_action
  parser.print_help()
    File "/usr/lib/python3.6/optparse.py", line 1647, in print_help
  file.write(self.format_help())
  UnicodeEncodeError: 'ascii' codec can't encode characters in position 
126-133: ordinal not in range(128)

  root@bb-lp-1806076:~# echo LANG=ru_RU.UTF-8 | tee /etc/default/locale
  LANG=ru_RU.UTF-8
  root@bb-lp-1806076:~# /usr/lib/apt/apt.systemd.daily update
  root@bb-lp-1806076:~#

  [Where problems could occur]

  * Nowhere, really. The fix is setting LC_ALL=C.UTF-8 for u-u --help
  and it is processed by grep then.

  [Original Bug Text]

  The Ubuntu Error Tracker has been receiving reports about a problem regarding 
unattended-upgrades.  This problem was most recently seen with package version 
1.1ubuntu1.18.04.6, the problem page at 
https://errors.ubuntu.com/problem/b3e3265e302351558260f54ae37c7b4c193dfc95 
contains more details, including versions of packages affected, stacktrace or 
traceback, and individual crash reports.
  If you do not have access to the Ubuntu Error Tracker and are a software 
developer, you can request it at http://forms.canonical.com/reports/.

  Also seen in:
   * https://errors.ubuntu.com/problem/936bb1c75c4efe018f968a5773b820bcf52c298a

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1806076/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1806076] Re: unattended-upgrade --help raises UnicodeEncodeError when stdout encoding is ascii

2021-03-30 Thread Julian Andres Klode
Log:

+ /usr/lib/apt/apt.systemd.daily lock_is_held update



--- Logging error ---   



Traceback (most recent call last):  



  File "/usr/lib/python3.6/logging/__init__.py", line 996, in emit  



stream.write(msg)   



UnicodeEncodeError: 'ascii' codec can't encode characters in position 29-34: 
ordinal not in range(128)   

   
Call stack: 



  File "/usr/bin/unattended-upgrade", line 2193, in 



sys.exit(main(options)) 



  File "/usr/bin/unattended-upgrade", line 1750, in main



" ".join(blacklisted_pkgs)) 



Message: '\u041f\u0430\u043a\u0435\u0442\u044b \u0432 
\u0447\u0451\u0440\u043d\u043e\u043c \u0441\u043f\u0438\u0441\u043a\u0435 
\u0438\u0437\u043d\u0430\u0447\u0430\u043b\u044c\u043d\u043e: %s'   

Arguments: ('',)



--- Logging error ---   



Traceback (most recent call last):  



  File "/usr/lib/python3.6/logging/__init__.py", line 996, in emit  



stream.write(msg)   



UnicodeEncodeError: 'ascii' codec can't encode characters in position 29-39: 
ordinal not in range(128)  

[Touch-packages] [Bug 1898026] Re: interruption of dist-upgrade can leave you next release in sources.list

2021-03-30 Thread Julian Andres Klode
I checked do-release-upgrade -d before upgrading apt, and pressing
Ctrl+C at the prompt for conforming changes left hirsute in
sources.list.

I upgraded to apt 2.1.10ubuntu0.3 from proposed, ran it again, and
Ctrl+C now caused the restore code to run:

 Continue [yN]  Details [d]^C
Restoring original system state

Aborting
Reading package lists... Done
Building dependency tree  
Reading state information... Done
=== Command detached from window (Tue Mar 30 15:04:31 2021) ===

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1898026

Title:
  interruption of dist-upgrade can leave you next release in
  sources.list

Status in apt package in Ubuntu:
  Fix Released
Status in ubuntu-release-upgrader package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  New
Status in ubuntu-release-upgrader source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in ubuntu-release-upgrader source package in Focal:
  New
Status in apt source package in Groovy:
  Fix Committed
Status in ubuntu-release-upgrader source package in Groovy:
  New
Status in apt source package in Hirsute:
  Fix Released
Status in ubuntu-release-upgrader source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  Calling update() or installing packages from apt clients resets their SIGINT 
and SIGQUIT handlers to SIG_DFL, overriding any signal handlers they might have 
set for them.

  In case of ubuntu-release-upgrader, this results in the release
  upgrader being unable to handle interrupts after it did the initial
  update - the default libc handler will run and the program exits.

  [Test plan]

  [[apt]]
  As a standalone test for apt, we can test the following script:

  import apt
  import time

  apt.Cache().update()
  print("WAITING")
  try:
  time.sleep(1)
  except BaseException as e:
  print("Seen", repr(e))
  print("END")

  Pressing Ctrl+C while WAITING is printed should print Seen
  KeyboardInterrupt, and importantly, also the END line.

  [Where problems could occur]
  apt: This specific change removes the two lines that SIG_DFL the signal 
handlers after running scripts. AFAWCT those lines are unnecessary - the code 
that calls it temporarily sets the handlers to SIG_IGN but restores previous 
handlers at the end; it was wrongly refactored decades ago. A regression could 
occur in that those signals will now continue to be ignored if we missed a spot.

  [Original bug report]
  As a long-time-user of ?ubuntu, with apt-get as tool of choice for 
updates/upgrades I wrote a daily script for updates, with 'dist-upgrade'.
  In all earlier years, it wouldn't actually do an upgrade of a ?ubuntu 
version; just all packages including new ones. Version updates had to be 
initiated manually, and I was always asked if I really wanted the new ?ubuntu 
version. Sounds appropriate.

  Last night when it (dist-upgrade), it just gave me 20.04. No questions asked. 
I for one consider this kind of intrusive, though.
  It *might* have to make with me trying 'sudo do-release-upgrade -m desktop' a 
number of times earlier; just to *check* if the upgrade was on offer; but this 
is only a guess.

  In *any* case, a pop-up asking "Are you sure? Are you connected
  through an adequate pipe? Are you sitting with full batteries; better
  a power supply?" would be convenient; since I am using my PC for
  urgent duties, and didn't want to fiddle with unexpected upgrade bugs.

  ProblemType: Bug
  DistroRelease: Ubuntu 20.04
  Package: ubuntu-release-upgrader-core 1:20.04.25
  ProcVersionSignature: Ubuntu 5.4.0-48.52-generic 5.4.60
  Uname: Linux 5.4.0-48-generic x86_64
  ApportVersion: 2.20.11-0ubuntu27.9
  Architecture: amd64
  CasperMD5CheckResult: skip
  CrashDB: ubuntu
  CurrentDesktop: KDE
  Date: Thu Oct  1 11:48:39 2020
  InstallationDate: Installed on 2019-03-14 (566 days ago)
  InstallationMedia: Kubuntu 18.04.1 LTS "Bionic Beaver" - Release amd64 
(20180725)
  PackageArchitecture: all
  SourcePackage: ubuntu-release-upgrader
  Symptom: ubuntu-release-upgrader
  UpgradeStatus: Upgraded to focal on 2020-09-29 (1 days ago)
  VarLogDistupgradeAptlog:
   Log time: 2020-09-29 19:44:28.696289
   Log time: 2020-09-29 19:44:33.573481

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1898026/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1898026] Re: interruption of dist-upgrade can leave you next release in sources.list

2021-03-30 Thread Julian Andres Klode
The small simple script also worked with the proposed version:

# python3 a.py
WAITING
^CSeen KeyboardInterrupt()
END


** Tags removed: verification-needed verification-needed-groovy
** Tags added: verification-done verification-done-groovy

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1898026

Title:
  interruption of dist-upgrade can leave you next release in
  sources.list

Status in apt package in Ubuntu:
  Fix Released
Status in ubuntu-release-upgrader package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  New
Status in ubuntu-release-upgrader source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in ubuntu-release-upgrader source package in Focal:
  New
Status in apt source package in Groovy:
  Fix Committed
Status in ubuntu-release-upgrader source package in Groovy:
  New
Status in apt source package in Hirsute:
  Fix Released
Status in ubuntu-release-upgrader source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  Calling update() or installing packages from apt clients resets their SIGINT 
and SIGQUIT handlers to SIG_DFL, overriding any signal handlers they might have 
set for them.

  In case of ubuntu-release-upgrader, this results in the release
  upgrader being unable to handle interrupts after it did the initial
  update - the default libc handler will run and the program exits.

  [Test plan]

  [[apt]]
  As a standalone test for apt, we can test the following script:

  import apt
  import time

  apt.Cache().update()
  print("WAITING")
  try:
  time.sleep(1)
  except BaseException as e:
  print("Seen", repr(e))
  print("END")

  Pressing Ctrl+C while WAITING is printed should print Seen
  KeyboardInterrupt, and importantly, also the END line.

  [Where problems could occur]
  apt: This specific change removes the two lines that SIG_DFL the signal 
handlers after running scripts. AFAWCT those lines are unnecessary - the code 
that calls it temporarily sets the handlers to SIG_IGN but restores previous 
handlers at the end; it was wrongly refactored decades ago. A regression could 
occur in that those signals will now continue to be ignored if we missed a spot.

  [Original bug report]
  As a long-time-user of ?ubuntu, with apt-get as tool of choice for 
updates/upgrades I wrote a daily script for updates, with 'dist-upgrade'.
  In all earlier years, it wouldn't actually do an upgrade of a ?ubuntu 
version; just all packages including new ones. Version updates had to be 
initiated manually, and I was always asked if I really wanted the new ?ubuntu 
version. Sounds appropriate.

  Last night when it (dist-upgrade), it just gave me 20.04. No questions asked. 
I for one consider this kind of intrusive, though.
  It *might* have to make with me trying 'sudo do-release-upgrade -m desktop' a 
number of times earlier; just to *check* if the upgrade was on offer; but this 
is only a guess.

  In *any* case, a pop-up asking "Are you sure? Are you connected
  through an adequate pipe? Are you sitting with full batteries; better
  a power supply?" would be convenient; since I am using my PC for
  urgent duties, and didn't want to fiddle with unexpected upgrade bugs.

  ProblemType: Bug
  DistroRelease: Ubuntu 20.04
  Package: ubuntu-release-upgrader-core 1:20.04.25
  ProcVersionSignature: Ubuntu 5.4.0-48.52-generic 5.4.60
  Uname: Linux 5.4.0-48-generic x86_64
  ApportVersion: 2.20.11-0ubuntu27.9
  Architecture: amd64
  CasperMD5CheckResult: skip
  CrashDB: ubuntu
  CurrentDesktop: KDE
  Date: Thu Oct  1 11:48:39 2020
  InstallationDate: Installed on 2019-03-14 (566 days ago)
  InstallationMedia: Kubuntu 18.04.1 LTS "Bionic Beaver" - Release amd64 
(20180725)
  PackageArchitecture: all
  SourcePackage: ubuntu-release-upgrader
  Symptom: ubuntu-release-upgrader
  UpgradeStatus: Upgraded to focal on 2020-09-29 (1 days ago)
  VarLogDistupgradeAptlog:
   Log time: 2020-09-29 19:44:28.696289
   Log time: 2020-09-29 19:44:33.573481

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1898026/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1877987] Re: apt search coredumping

2021-03-30 Thread Julian Andres Klode
Can't exactly verify as we did not get another reproducer, but marking
it as verified anyway

** Tags removed: verification-needed verification-needed-focal
** Tags added: verification-done verification-done-focal

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1877987

Title:
  apt search coredumping

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Focal:
  Fix Committed

Bug description:
  [Impact]
  Search sometimes crashes if no translated description is available, as the 
translated description object was used without a check as to whether it exists.

  [Test case]
  No reproducer available at the moment.

  
  [Where problems could occur]
  In search command only, and only when looking at descriptions

  [Original bug report]
  root@citoyx:/var/crash# apt search jdk
  Sorting... Done
  Segmentation fault (core dumped)

  May 11 12:34:25 citoyx kernel: [ 1354.337907] apt[10418]: segfault at 
7f67292dcfc0 ip 7f6309ce4997 sp 7ffdab880930 error 4 in 
libapt-pkg.so.6.0.0[7f6309bb5000+147000]
  May 11 12:34:25 citoyx kernel: [ 1354.337916] Code: 48 8b 45 18 48 8b 04 d0 
48 83 c4 08 5b 5d c3 0f 1f 40 00 f3 0f 1e fa 55 48 89 fd 53 48 89 f3 48 83 ec 
08 48 8b 06 48 8b 4e 08 <8b> 10 48 8d 04 d5 00 00 00 00 48 29 d0 48 8b 51 60 48 
8d 04 c2 0f

  root@citoyx:/var/crash# cat /etc/lsb-release
  DISTRIB_ID=Ubuntu
  DISTRIB_RELEASE=20.04
  DISTRIB_CODENAME=focal
  DISTRIB_DESCRIPTION="Ubuntu 20.04 LTS"

  apt search is coredumping
  While apt install/upgrade/update is running fine

  crash file in attachment.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1877987/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1916725] Re: Protected/Important packages are not deconfigured, require Force-LoopBreak

2021-03-30 Thread Julian Andres Klode
** Tags removed: verification-donefocal
** Tags added: verification-done-focal

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1916725

Title:
  Protected/Important packages are not deconfigured, require Force-
  LoopBreak

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  If a package that is Protected: yes (or Important: yes), or one of it's 
dependencies, is involved in a dependency loop with Breaks, APT requires 
APT::Force-LoopBreak instead of resolving the situation directly.

  On focal, we also introduce the actual support for protected packages
  to enable upgrading to later releases more easily (in case a protected
  package needs to be removed during the upgrade), and to make the
  backport more similar to main.

  [Test plan]
  Run the integration test suite (the autopkgtest) :)

  Our test suite covers the tests for both Breaks and Conflicts.

  Breaks:

  protected-sysvinit (= 1) without dependencies is installed
  protected-sysvinit (= 2) Pre-Depends protected-systemd-sysv
  protected-systemd-sysv (= 2) Breaks: protected-sysvinit (<< 2)

  Test: Install protected-sysvinit (= 2)
  Expected result: Unpacking protected-sysvinit (= 2) deconfigures 
protected-sysvinit (= 1), and then we unpack and configure protected-sysvinit 
(= 2) and end up with a working system.

  Conflicts: As for Breaks, but the Conflicts will remove the package
  temporarily, requiring the use of APT::Force-LoopBreak option.

  For focal, we also do have a test to check that the Protected field is
  being used.

  [Where problems could occur]
  We now allow dpkg to automatically deconfigure protected packages. This 
should just make them behave like normal packages to APT's eye, but bugs I 
guess could occur somewhere in the APT/dpkg interaction (this only applies to 
releases with Protected support in dpkg, Important is not affected, it's always 
been "normal" for dpkg).

  During development, we accidentally simplified the patch so much that
  Conflicts did not require Force-LoopBreak for temporary removal. We
  fixed that, but it points out that there is a place where the loop
  break check happens that is a potential regression place.

  On focal, we pass additional flags to dpkg that focal's dpkg does not
  understand, however, we only do that if dpkg asserts it does that, so
  in practice, this should all work fine and the code path will only be
  taken with >=groovy dpkg.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1916725/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918907] Re: Default Acquire::AllowReleaseInfoChange::Suite to "true"

2021-03-30 Thread Julian Andres Klode
Oops, of course apt's test passed, hence verification done :)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918907

Title:
  Default Acquire::AllowReleaseInfoChange::Suite to "true"

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed

Bug description:
  [Impact]
  APT 1.6-2.0 deny third-party repositories from changing their Suite value. 
This change was reverted in later apt releases, and will be reverted in 1.8 in 
Debian as well, so for consistency, we want to push that change to stable

  [Test plan]
  We added test cases to our extensive integration test suite run by 
autopkgtest.

  [Where problems could occur]
  It's hard to imagine anything here, we simply turn the flag for allowed from 
false to true. We even add tests for Suite changes :)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918907/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918907] Re: Default Acquire::AllowReleaseInfoChange::Suite to "true"

2021-03-30 Thread Julian Andres Klode
apt's tests failed, I believe the regressions are just intermittent
unrelated issues.

** Tags removed: verification-needed verification-needed-bionic 
verification-needed-focal
** Tags added: verification-done verification-done-bionic 
verification-done-focal

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918907

Title:
  Default Acquire::AllowReleaseInfoChange::Suite to "true"

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed

Bug description:
  [Impact]
  APT 1.6-2.0 deny third-party repositories from changing their Suite value. 
This change was reverted in later apt releases, and will be reverted in 1.8 in 
Debian as well, so for consistency, we want to push that change to stable

  [Test plan]
  We added test cases to our extensive integration test suite run by 
autopkgtest.

  [Where problems could occur]
  It's hard to imagine anything here, we simply turn the flag for allowed from 
false to true. We even add tests for Suite changes :)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918907/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1907850] Re: Cache not generated for all translations

2021-03-30 Thread Julian Andres Klode
autopkgtests for apt have passed, so validation was successful. I'm
confident the autopkgtest failures are unrelated intermittent ones; I've
done some more retries.

** Tags removed: verification-needed verification-needed-bionic 
verification-needed-focal verification-needed-groovy
** Tags added: verification-done verification-done-bionic 
verification-done-focal verification-done-groovy

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1907850

Title:
  Cache not generated for all translations

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Confirmed
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  In bug 1161743 we discovered that if a system is configured with multiple 
locales, only the locales of the user who generated the apt-cache will be 
available for translated descriptions.

  [Test case]
  An automated test case has been included as a snippet in  
test/integration/test-bug-49-search-in-all-translations which runs during 
autopkgtest. That's sufficient for validation.

  Manual test:

  # apt install locales-all # get the locale
  # export LANG=sv_SE.UTF-8
  # locale
  LANG=sv_SE.UTF-8
  LANGUAGE=
  LC_CTYPE="sv_SE.UTF-8"
  LC_NUMERIC="sv_SE.UTF-8"
  LC_TIME="sv_SE.UTF-8"
  LC_COLLATE="sv_SE.UTF-8"
  LC_MONETARY="sv_SE.UTF-8"
  LC_MESSAGES="sv_SE.UTF-8"
  LC_PAPER="sv_SE.UTF-8"
  LC_NAME="sv_SE.UTF-8"
  LC_ADDRESS="sv_SE.UTF-8"
  LC_TELEPHONE="sv_SE.UTF-8"
  LC_MEASUREMENT="sv_SE.UTF-8"
  LC_IDENTIFICATION="sv_SE.UTF-8"
  LC_ALL=
  # apt update
  # apt-cache show tasksel | grep Desc
  Description-sv: tool for selecting tasks for installation on Debian systems
  Description-md5: cbbb747708986d11ea77c80b9b038fec
  # apt-cache showpkg tasksel
  Package: tasksel
  Versions:
  3.34ubuntu16 
(/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages)
   Description Language:
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages
    MD5: cbbb747708986d11ea77c80b9b038fec
   Description Language: sv
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_i18n_Translation-sv
    MD5: cbbb747708986d11ea77c80b9b038fec
   Description Language: en
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_i18n_Translation-en
    MD5: cbbb747708986d11ea77c80b9b038fec
  [...]

  So far so good, but now assume the root user actually has C configured
  as locale, and e.g. runs apt-cache show (or apt-daily.service does an
  update):

  root@g:~# rm /var/cache/apt/*.bin
  root@g:~# LANG=C apt-cache  show tasksel
  [...]
  Description-en: tool for selecting tasks for installation on Debian systems
   This package provides 'tasksel', a simple interface for users who
   want to configure their system to perform a specific task.

  root@g:~# apt-cache  showpkg tasksel
  Package: tasksel
  Versions:
  3.34ubuntu16 
(/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages)
   Description Language:
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages
    MD5: cbbb747708986d11ea77c80b9b038fec
   Description Language: en
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_i18n_Translation-en
    MD5: cbbb747708986d11ea77c80b9b038fec

  This should show the sv locale as well given that it's still around
  (also we are still running with LANG=sv_SE.UTF-8), but it only
  generated the cache with the english language description in here.

  [Where problems could occur]
  People might end up seeing a wrong translation if other code paths are wrong, 
but then they'd already see them if they explicitly configured multiple 
languages and then switched the preferred one.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1907850/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1916050] Re: Invalid base64 for high-bit characters

2021-03-30 Thread Julian Andres Klode
autopkgtests for apt have passed, so validation was successful. I'm
confident the autopkgtest failures are unrelated intermittent ones; I've
done some more retries.

** Tags removed: verification-needed verification-needed-bionic 
verification-needed-focal verification-needed-groovy
** Tags added: verification-done verification-done-bionic 
verification-done-focal verification-done-groovy

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1916050

Title:
  Invalid base64 for high-bit characters

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Confirmed
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  The Base64Encode function incorrectly encodes using char instead of uint8_t, 
leading to characters with the high bit set to be encoded wrongly. This 
function is used for base authentication, and high bits are set for UTF-8 
character sequences, so it's likely affecting people with utf-8 user names and 
passwords and might cause them to be denied access to their https resources 
(though arguably they'd work around that).

  [Test case]
  A unit test has been added that checks that the encoding is correct.

  [Where problems could occur]
  This only affects the base64 encoding function, which is only used for 
quoting user names and passwords in basic auth in http, but it's also exposed 
to library users and like in python-apt, so problems could occur wherever they 
use that base64 encoding function as well.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1916050/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1916725] Re: Protected/Important packages are not deconfigured, require Force-LoopBreak

2021-03-30 Thread Julian Andres Klode
autopkgtests for apt have passed, so validation was successful. I'm
confident the autopkgtest failures are unrelated intermittent ones; I've
done some more retries.

** Tags removed: verification-needed verification-needed-bionic 
verification-needed-focal verification-needed-groovy
** Tags added: verification-done verification-done-bionic 
verification-done-groovy verification-donefocal

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1916725

Title:
  Protected/Important packages are not deconfigured, require Force-
  LoopBreak

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  If a package that is Protected: yes (or Important: yes), or one of it's 
dependencies, is involved in a dependency loop with Breaks, APT requires 
APT::Force-LoopBreak instead of resolving the situation directly.

  On focal, we also introduce the actual support for protected packages
  to enable upgrading to later releases more easily (in case a protected
  package needs to be removed during the upgrade), and to make the
  backport more similar to main.

  [Test plan]
  Run the integration test suite (the autopkgtest) :)

  Our test suite covers the tests for both Breaks and Conflicts.

  Breaks:

  protected-sysvinit (= 1) without dependencies is installed
  protected-sysvinit (= 2) Pre-Depends protected-systemd-sysv
  protected-systemd-sysv (= 2) Breaks: protected-sysvinit (<< 2)

  Test: Install protected-sysvinit (= 2)
  Expected result: Unpacking protected-sysvinit (= 2) deconfigures 
protected-sysvinit (= 1), and then we unpack and configure protected-sysvinit 
(= 2) and end up with a working system.

  Conflicts: As for Breaks, but the Conflicts will remove the package
  temporarily, requiring the use of APT::Force-LoopBreak option.

  For focal, we also do have a test to check that the Protected field is
  being used.

  [Where problems could occur]
  We now allow dpkg to automatically deconfigure protected packages. This 
should just make them behave like normal packages to APT's eye, but bugs I 
guess could occur somewhere in the APT/dpkg interaction (this only applies to 
releases with Protected support in dpkg, Important is not affected, it's always 
been "normal" for dpkg).

  During development, we accidentally simplified the patch so much that
  Conflicts did not require Force-LoopBreak for temporary removal. We
  fixed that, but it points out that there is a place where the loop
  break check happens that is a potential regression place.

  On focal, we pass additional flags to dpkg that focal's dpkg does not
  understand, however, we only do that if dpkg asserts it does that, so
  in practice, this should all work fine and the code path will only be
  taken with >=groovy dpkg.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1916725/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918920] Re: Harden test for no new acquires after transaction abort

2021-03-30 Thread Julian Andres Klode
autopkgtests for apt have passed, so validation was successful. I'm
confident the autopkgtest failures are unrelated intermittent ones; I've
done some more retries.

** Tags removed: verification-needed verification-needed-bionic 
verification-needed-focal verification-needed-groovy
** Tags added: verification-done verification-done-bionic 
verification-done-focal verification-done-groovy

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918920

Title:
  Harden test for no new acquires after transaction abort

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed

Bug description:
  [Impact]
  test-pdiff-usage is somewhat flaky, especially on Debian, this makes it less 
flaky. No end user impact as it's a test-only change.

  [Test plan]
  Running autopkgtest, which runs our extensive integration test suite which 
includes the changed test.

  [Where problems could occur]
  No end user regression potential on its own, but might slightly change 
regression potential for future pdiff changes:

  Test approach is slightly different now. It still catches that updates
  fail correctly, but tests more concretely that a transaction was
  aborted rather than that no worker received work (which was not
  guaranteed, the work could be scheduled before it was aborted).

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918920/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1615381] Re: apt-get autoremove may remove current kernel

2021-03-30 Thread Julian Andres Klode
The autopkgtests for apt have passed, so testing succeeded. I retried
the failures, I expect them to be unrelated: apport one already happened
with binutils/2.30-21ubuntu1~18.04.5, the i386 retry succeeded; and
reprotest fails in its normal way, a few retries should get it working.

** Tags removed: verification-needed verification-needed-bionic 
verification-needed-focal verification-needed-groovy
** Tags added: verification-done verification-done-bionic 
verification-done-focal verification-done-groovy

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to unattended-upgrades in
Ubuntu.
https://bugs.launchpad.net/bugs/1615381

Title:
  apt-get autoremove may remove current kernel

Status in apt package in Ubuntu:
  Fix Released
Status in unattended-upgrades package in Ubuntu:
  Fix Released
Status in unattended-upgrades source package in Trusty:
  Won't Fix
Status in unattended-upgrades source package in Xenial:
  Fix Released
Status in unattended-upgrades source package in Artful:
  Won't Fix
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  APT will try and fail to remove the currently running kernel, when booted 
into an older kernel that is not protected. May wreak some havoc if mixed with 
other operations to the point that apt goes weird and needs manual fixing up.

  [Test case]
  For the SRUs we have included an automated test case that starts with an 
empty autoremove config file, and then checks that the running kernel is 
protected at runtime.

  You can also test manually, but it's not necessary:

  - Install new kernel
  - Reboot into kernel not listed as protected in 01autoremove-kernels
  - Run autoremove

  [Where problems could occur]
  We may more easily run out of space in /boot. hirsute has new autoremoval 
code that runs completely at runtime; but that seems a bit large to SRU after 
only a few weeks in hirsute. Hence, we should protect the current kernel _in 
addition_ to the other kernels, just like unattended-upgrades and 
update-manager do. This increases the risk of filling up /boot compared to 
older apt versions, but is at the same level as unattended-upgrades and 
update-manager.

  [Original bug report]

  This may happen, if you boot one of the older kernels, that is not
  protected by /etc/apt/apt.conf.d/01autoremove-kernels

  Workaround: run
  /etc/kernel/postinst.d/apt-auto-removal
  during each boot (e.g. by using cron).
  Note: The workaround breaks autoremoving feature of new unneeded kernels in  
unattended-upgrades i.e. the setting 
'Unattended-Upgrade::Remove-New-Unused-Dependencies "true"' (which is default 
in 16.04 unless 'Unattended-Upgrade::Remove-Unused-Dependencies "true"' is set 
in '/etc/apt/apt.conf.d/50unattended-upgrades'.

  In shell:

  $ uname -r
  4.4.0-22-generic
  $ apt-get -s autoremove
  NOTE: This is only a simulation!
    apt-get needs root privileges for real execution.
    Keep also in mind that locking is deactivated,
    so don't depend on the relevance to the real current situation!
  Reading package lists... Done
  Building dependency tree
  Reading state information... Done
  The following packages will be REMOVED:
    linux-headers-4.4.0-21 linux-headers-4.4.0-21-generic linux-headers-4.4.0-22
    linux-headers-4.4.0-22-generic linux-headers-4.4.0-31-generic
    linux-image-4.4.0-21-generic linux-image-4.4.0-22-generic
    linux-image-4.4.0-31-generic linux-image-extra-4.4.0-21-generic
    linux-image-extra-4.4.0-22-generic linux-image-extra-4.4.0-31-generic
  0 upgraded, 0 newly installed, 11 to remove and 13 not upgraded.

  ProblemType: Bug
  DistroRelease: Ubuntu 16.04
  Package: apt 1.2.12~ubuntu16.04.1
  ProcVersionSignature: Ubuntu 4.4.0-22.40-generic 4.4.8
  Uname: Linux 4.4.0-22-generic x86_64
  ApportVersion: 2.20.1-0ubuntu2.1
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sun Aug 21 16:11:27 2016
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2016-04-28 (114 days ago)
  InstallationMedia: Xubuntu 16.04 LTS "Xenial Xerus" - Release amd64 
(20160420.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.kernel.postinst.d.apt-auto-removal: [modified]
  mtime.conffile..etc.kernel.postinst.d.apt-auto-removal: 
2016-07-30T12:15:32.706300

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1615381/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1693900] Re: apt-get update should return exit code != 0 on error

2021-03-30 Thread Julian Andres Klode
The autopkgtests for apt have passed, so testing succeeded. I retried
the failures, I expect them to be unrelated: apport one already happened
with binutils/2.30-21ubuntu1~18.04.5, the i386 retry succeeded; and
reprotest fails in its normal way, a few retries should get it working.


** Tags removed: verification-needed-bionic verification-needed-focal 
verification-needed-groovy
** Tags added: verification-done-bionic verification-done-focal 
verification-done-groovy

** Tags removed: verification-needed
** Tags added: verification-done

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1693900

Title:
  apt-get update should return exit code != 0 on error

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Confirmed
Status in apt source package in Bionic:
  Fix Committed
Status in apt source package in Focal:
  Fix Committed
Status in apt source package in Groovy:
  Fix Committed
Status in apt package in Debian:
  Fix Released

Bug description:
  [Impact]
  There is no way for scripts to run apt update and be sure that it was 
entirely successful.

  We introduce a new flag, --error-on=any that makes apt update also
  error out on transient errors.

  [Test plan]
  We have included a test in the test-suite in 
test-apt-update-failure-propagation that ensures that warnings become errors.

  [Where problems could occur]
  If there's an issue some warnings for transient errors could become errors, 
or worse I suppose it's possible but unlikely for errors to be reported as 
warnings...

  Since the new behavior is behind a flag, regression potential for
  existing scripts is low if there is no bug.

  [Original bug report]

  When running 'apt-get update' (e.g. on a container install post-
  install script), apt-get return with exit code 0, even so it wasn't
  able to "update" properly. E.g.:

  + apt-get update
  Err:1 http://de.archive.ubuntu.com/ubuntu xenial InRelease
    Temporary failure resolving 'de.archive.ubuntu.com'
  Err:2 http://security.ubuntu.com/ubuntu xenial-security InRelease
    Temporary failure resolving 'security.ubuntu.com'
  Err:3 http://de.archive.ubuntu.com/ubuntu xenial-updates InRelease
    Temporary failure resolving 'de.archive.ubuntu.com'
  Reading package lists... Done
  W: Failed to fetch http://de.archive.ubuntu.com/ubuntu/dists/xenial/InRelease 
 Temporary failure resolving 'de.archive.ubuntu.com'
  W: Failed to fetch 
http://de.archive.ubuntu.com/ubuntu/dists/xenial-updates/InRelease  Temporary 
failure resolving 'de.archive.ubuntu.com'
  W: Failed to fetch 
http://security.ubuntu.com/ubuntu/dists/xenial-security/InRelease  Temporary 
failure resolving 'security.ubuntu.com'
  W: Some index files failed to download. They have been ignored, or old ones 
used instead.

  It should be corrected to return useful exit code, so that scripts can
  take the appropriate actions ...

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1693900/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921773] Re: apt upgrade clears a hold on an uninstalled snapd

2021-03-29 Thread Julian Andres Klode
This was fixed in  dpkg (1.20.6) unstable; urgency=medium

   * libdpkg: Do not forget not-installed packages that are set on hold.
 Reported by David Kalnischkies .

** Changed in: apt (Ubuntu)
   Status: New => Fix Released

** Package changed: apt (Ubuntu) => dpkg (Ubuntu)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921773

Title:
  apt upgrade clears a hold on an uninstalled snapd

Status in dpkg package in Ubuntu:
  Fix Released

Bug description:
  I have an "apt-mark hold" on several packages, one of which is not
  installed, snapd. An "apt upgrade" that's not a noop, that actually
  installs something, clears the snapd hold. The others remain.

  ProblemType: Bug
  DistroRelease: Ubuntu 20.04
  Package: apt 2.0.4
  ProcVersionSignature: Ubuntu 5.8.0-48.54~20.04.1-generic 5.8.18
  Uname: Linux 5.8.0-48-generic x86_64
  ApportVersion: 2.20.11-0ubuntu27.16
  Architecture: amd64
  CasperMD5CheckResult: skip
  CurrentDesktop: ubuntu:GNOME
  Date: Mon Mar 29 12:47:58 2021
  InstallationDate: Installed on 2020-10-16 (163 days ago)
  InstallationMedia: Ubuntu 20.04.1 LTS "Focal Fossa" - Release amd64 (20200731)
  ProcEnviron:
   TERM=xterm-256color
   PATH=(custom, no user)
   XDG_RUNTIME_DIR=
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/dpkg/+bug/1921773/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921626] Re: apt install - File has unexpected size - http pipeline

2021-03-29 Thread Julian Andres Klode
We tracked this down to the Pulp repository not having Size fields.
While this is a broken repository, APT should also more gracefully
handle that situation, either by warning about missing Size values,
erroring out, disabling pipelining; or by fixing pipelining to behave if
we don't know the size in advance.

** Changed in: apt (Ubuntu)
   Status: Incomplete => Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  apt install - File has unexpected size - http pipeline

Status in apt package in Ubuntu:
  Triaged

Bug description:
  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  
  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921626] Re: apt install - File has unexpected size - http pipeline

2021-03-29 Thread Julian Andres Klode
> I don't have that mirrored, but I have focal, and I could not
reproduce. Apt 2.0.2

There have been no relevant changes between 1.6.y and 2.0.y, so it
really should reproduce with both of them or none of them.

But you have to be careful that the download is _exactly_ the same,
usually by pointing new apt at the older system state, otherwise you
might have lost your reproducer.

But yeah, this "comes and goes" is why we're not making much progress on
HTTP bugs.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  apt install - File has unexpected size - http pipeline

Status in apt package in Ubuntu:
  Incomplete

Bug description:
  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  
  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921626] Re: apt install - File has unexpected size - http pipeline

2021-03-28 Thread Julian Andres Klode
Managing to get a reproducer for HTTP issues more than a couple of hours
has not really worked so far, and most people's reproducers don't work.
It's highly subjective to network latency and MTU afaict.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  apt install - File has unexpected size - http pipeline

Status in apt package in Ubuntu:
  Incomplete

Bug description:
  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  
  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1921626] Re: apt install - File has unexpected size - http pipeline

2021-03-28 Thread Julian Andres Klode
Please let me know if you can reproduce this with 2.2.2 in hirsute. No
changes can be made to the http method in stable releases, as it's super
fragile - Every time we make changes, something else breaks.

** Changed in: apt (Ubuntu)
   Status: New => Incomplete

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1921626

Title:
  apt install - File has unexpected size - http pipeline

Status in apt package in Ubuntu:
  Incomplete

Bug description:
  1) Ubuntu 18.04.5 LTS

  2) apt 1.6.12ubuntu0.2

  3) What you expected to happen

  I set a custom set of repositories in /etc/apt/sources.list and then I
  run "apt install ". I expect the command to download
  and install the packages.

  4) What happened instead

  "apt install ..." fails during the download phase with "File has
  unexpected size "

  
  5) What I've established trying to debug the issue:

  - Disabling http pipelining resolves the issue: "apt 
-oAcquire::http::Pipeline-Depth=0 install ..."
  - All the packages, and repo metadata in the referenced repositories is 
correct
  - The issue is easily reproducible in my setup with different repositories
  - tcpdump shows that requests and responses are in the correct order, and 
contain the correct data

  More details about the issue:
  https://projects.theforeman.org/issues/32178

  With all the above in mind, it appears that this must be a bug in apt's http 
pipeline handling.
  It seem that apt is trying to match a request to do wrong response, and size 
doesn't match. 

  I've attached an example log, where the error pops up for multiple
  packages, and they all appear to be compared to one size (86464
  bytes). That size is correct for one of the package being downloaded,
  but somehow apt is trying to match to multiple other packages.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1921626/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 451366] Re: 404 from the archive is not handled well

2021-03-26 Thread Julian Andres Klode
Mathew, please don't mark unrelated bugs as duplicates.

** This bug is no longer a duplicate of bug 1693900
   apt-get update should return exit code != 0 on error

** Changed in: apt (Ubuntu)
   Importance: Undecided => Wishlist

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/451366

Title:
  404 from the archive is not handled well

Status in Aptdaemon:
  In Progress
Status in apt package in Ubuntu:
  New
Status in aptdaemon package in Ubuntu:
  Triaged
Status in software-center package in Ubuntu:
  Triaged

Bug description:
  Binary package hint: software-center

  When trying to install Pingus from the Ubuntu Software Center It
  downloads to 50% and then gives the following error message:

  Failed to download package files
  Check your internet connection.
  Details
  Failed to fetch 
http://us.archive.ubuntu.com/ubuntu/pool/main/b/boost1.38/libboost-signals1.38.0_1.38.0-6ubuntu5_i386.deb
 404  Not Found [IP: 91.189.88.31 80]


  My internet is working and I have downloaded and installed other
  software succesfully from the software center.

  ProblemType: Bug
  Architecture: i386
  Date: Wed Oct 14 08:20:35 2009
  DistroRelease: Ubuntu 9.10
  Package: software-center 0.4.6
  PackageArchitecture: all
  ProcEnviron:
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  ProcVersionSignature: Ubuntu 2.6.31-12.39-generic
  SourcePackage: software-center
  Tags:  ubuntu-unr
  Uname: Linux 2.6.31-12-generic i686

To manage notifications about this bug go to:
https://bugs.launchpad.net/aptdaemon/+bug/451366/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1920601] Re: Frequent test failures caused by networking issues on armhf runners

2021-03-24 Thread Julian Andres Klode
We can set Acquire::Retries in apt.conf like Debian does for debci now

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1920601

Title:
  Frequent test failures caused by networking issues on armhf runners

Status in Auto Package Testing:
  Triaged
Status in apt package in Ubuntu:
  New

Bug description:
  Armhf runners have been known to have networking issues for years.

  A test of Focal's glibc SRU in Bileto shows that those issues did not
  go away. The glibc package itself is not expected to cause regression,
  this was just a run to confirm that.

  The test resuts are shown here:
  
https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_39a8dbb93caf4ec889f8a1b7f69885db/bileto-4017-excuses/2021-03-20_09:25:01/4017_focal_excuses.html

  Out of the 114 failed tests 58 containst the 'Connection timed out'
  string and only test logs from armhf contain that string:

  $ zgrep -l 'Connection timed out' ~/.cache/ubuntu-archive-tools/focal_*.gz 
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_atk1.0_20210318_182825_daefa.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_axe-demultiplexer_20210318_183430_6b008.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_bacula_20210318_190411_e3ad7.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_dialign-t_20210318_204503_b4d94.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_elisa-player_20210318_211129_9537d.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_fyba_20210318_181755_7731e.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gcc-7_20210318_02_4d95c.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gcc-8_20210318_221659_0e69f.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_golang-github-mailru-easyjson_20210318_231917_67888.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gphoto2_20210318_200015_eb767.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_gtk+3.0_20210318_235412_db3e5.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_ipmitool_20210319_003215_d03da.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_kbibtex_20210319_004639_0ac1a.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_kcptun_20210318_193309_6dcd9.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libcache-memcached-getparserxs-perl_20210319_024051_33029.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libclass-load-xs-perl_20210319_025846_01419.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libconvert-uulib-perl_20210319_034852_21cd1.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libhat-trie_20210319_060826_fd334.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libhtml-tidy5-perl_20210319_070723_05fa8.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libimager-perl_20210319_063314_b19a2.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libipc-sharelite-perl_20210319_072246_0217e.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_liblexical-var-perl_20210319_070235_4c450.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libmarpa-r2-perl_20210319_073731_f7554.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libmemcached-libmemcached-perl_20210319_011004_dbca1.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libnet-ldns-perl_20210319_081224_49912.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libnet-pcap-perl_20210319_081606_4d772.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libparams-util-perl_20210319_084831_74d13.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libperlio-eol-perl_20210319_085645_97c6c.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libperlio-layers-perl_20210319_085937_8bd70.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libsass_20210319_092518_4db37.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libsdl-perl_20210319_095502_00c8c.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libtext-bibtex-perl_20210319_105027_abf4a.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libtext-hunspell-perl_20210319_103326_4a603.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libunix-syslog-perl_20210319_111403_f4374.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libwx-scintilla-perl_20210319_114818_a6aab.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_libzstd_20210319_114737_0c4d9.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_mod-gearman_20210319_132205_d41ca.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_mrbayes_20210319_133633_10adb.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_octave-interval_20210319_151509_8348f.gz
  
/home/rbalint/.cache/ubuntu-archive-tools/focal_armhf_open-isns_20210319_161741_dffad.gz
  

[Touch-packages] [Bug 1713219] Re: 'apt-mark showauto' and 'apt show' is slow

2021-03-23 Thread Julian Andres Klode
There is no performance issue with the file format, the parsing is not
noticeable.

There is a massive design problem in the DepCache having to visit every
dependency in the cache and do marking stuff. This worked fine for tiny
sets, but scales superlinear to package count.

That's a very long time effort thing to fix for a future ABI break, we
need to only check those dependencies we actually care about.

Out of those 0.5s, probably about .4s are spent checking dependencies.
This will decrease by factor 10-50 presumably.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1713219

Title:
  'apt-mark showauto' and 'apt show' is slow

Status in apt package in Ubuntu:
  Triaged

Bug description:
  $ time apt-mark showauto >/dev/null

  real  0m0.587s
  user  0m0.552s
  sys   0m0.016s

  When I run the command first time, it is even much slower.

  I could do the job in fraction of a time using awk in POSIX shell
  script:

  auto_file='/var/lib/apt/extended_states'
  eval $(apt-config shell auto_file Dir::State::extended_states/f)
  awk '/^Package:/ {
pkg=$2
getline; arch=$2
getline
if($2==1) print pkg ":" arch
  }' "$auto_file" | CL_ALL=C sort -u

  real  0m0.019s
  user  0m0.008s
  sys   0m0.000s

  That prints architecture for every package and shows entries in
  slightly different order, though. And the file could be out of date
  showing packages that are not installed?!?

  Similarly

  apt show 

  is slow. (It also shows whether a package is manually or automatically
  installed.

  ProblemType: Bug
  DistroRelease: Ubuntu 14.04
  Package: apt 1.0.1ubuntu2.17
  ProcVersionSignature: Ubuntu 4.4.0-92.115~14.04.1-generic 4.4.76
  Uname: Linux 4.4.0-92-generic x86_64
  ApportVersion: 2.14.1-0ubuntu3.25
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sat Aug 26 12:59:00 2017
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2014-09-21 (1070 days ago)
  InstallationMedia: Ubuntu-Studio 14.04.1 LTS "Trusty Tahr" - Release amd64 
(20140722.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.cron.daily.apt: [modified]
  modified.conffile..etc.kernel.postinst.d.apt.auto.removal: [modified]
  mtime.conffile..etc.cron.daily.apt: 2017-05-03T10:27:27.617839
  mtime.conffile..etc.kernel.postinst.d.apt.auto.removal: 
2017-06-01T14:39:39.236080

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1713219/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1713219] Re: 'apt-mark showauto' and 'apt show' is slow

2021-03-23 Thread Julian Andres Klode
It's also imaginable that we might change the value from e.g. 1 to yes.
I don't know why it's an integer. In any case, I'd say you can't rely on
that file at all. It might get changed in format, renamed or removed
entirely without any notice.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1713219

Title:
  'apt-mark showauto' and 'apt show' is slow

Status in apt package in Ubuntu:
  Triaged

Bug description:
  $ time apt-mark showauto >/dev/null

  real  0m0.587s
  user  0m0.552s
  sys   0m0.016s

  When I run the command first time, it is even much slower.

  I could do the job in fraction of a time using awk in POSIX shell
  script:

  auto_file='/var/lib/apt/extended_states'
  eval $(apt-config shell auto_file Dir::State::extended_states/f)
  awk '/^Package:/ {
pkg=$2
getline; arch=$2
getline
if($2==1) print pkg ":" arch
  }' "$auto_file" | CL_ALL=C sort -u

  real  0m0.019s
  user  0m0.008s
  sys   0m0.000s

  That prints architecture for every package and shows entries in
  slightly different order, though. And the file could be out of date
  showing packages that are not installed?!?

  Similarly

  apt show 

  is slow. (It also shows whether a package is manually or automatically
  installed.

  ProblemType: Bug
  DistroRelease: Ubuntu 14.04
  Package: apt 1.0.1ubuntu2.17
  ProcVersionSignature: Ubuntu 4.4.0-92.115~14.04.1-generic 4.4.76
  Uname: Linux 4.4.0-92-generic x86_64
  ApportVersion: 2.14.1-0ubuntu3.25
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sat Aug 26 12:59:00 2017
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2014-09-21 (1070 days ago)
  InstallationMedia: Ubuntu-Studio 14.04.1 LTS "Trusty Tahr" - Release amd64 
(20140722.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.cron.daily.apt: [modified]
  modified.conffile..etc.kernel.postinst.d.apt.auto.removal: [modified]
  mtime.conffile..etc.cron.daily.apt: 2017-05-03T10:27:27.617839
  mtime.conffile..etc.kernel.postinst.d.apt.auto.removal: 
2017-06-01T14:39:39.236080

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1713219/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1713219] Re: 'apt-mark showauto' and 'apt show' is slow

2021-03-23 Thread Julian Andres Klode
It's slow because it loads the entire depcache, and checks all
dependencies. This can be avoided by refactoring the code, but it seems
unnecessary. I'd rather add useful features like telling you which
packages become autoremovable garbage after marking something, rather
than refactoring the code to not require a depcache.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1713219

Title:
  'apt-mark showauto' and 'apt show' is slow

Status in apt package in Ubuntu:
  Triaged

Bug description:
  $ time apt-mark showauto >/dev/null

  real  0m0.587s
  user  0m0.552s
  sys   0m0.016s

  When I run the command first time, it is even much slower.

  I could do the job in fraction of a time using awk in POSIX shell
  script:

  auto_file='/var/lib/apt/extended_states'
  eval $(apt-config shell auto_file Dir::State::extended_states/f)
  awk '/^Package:/ {
pkg=$2
getline; arch=$2
getline
if($2==1) print pkg ":" arch
  }' "$auto_file" | CL_ALL=C sort -u

  real  0m0.019s
  user  0m0.008s
  sys   0m0.000s

  That prints architecture for every package and shows entries in
  slightly different order, though. And the file could be out of date
  showing packages that are not installed?!?

  Similarly

  apt show 

  is slow. (It also shows whether a package is manually or automatically
  installed.

  ProblemType: Bug
  DistroRelease: Ubuntu 14.04
  Package: apt 1.0.1ubuntu2.17
  ProcVersionSignature: Ubuntu 4.4.0-92.115~14.04.1-generic 4.4.76
  Uname: Linux 4.4.0-92-generic x86_64
  ApportVersion: 2.14.1-0ubuntu3.25
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sat Aug 26 12:59:00 2017
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2014-09-21 (1070 days ago)
  InstallationMedia: Ubuntu-Studio 14.04.1 LTS "Trusty Tahr" - Release amd64 
(20140722.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.cron.daily.apt: [modified]
  modified.conffile..etc.kernel.postinst.d.apt.auto.removal: [modified]
  mtime.conffile..etc.cron.daily.apt: 2017-05-03T10:27:27.617839
  mtime.conffile..etc.kernel.postinst.d.apt.auto.removal: 
2017-06-01T14:39:39.236080

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1713219/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1713219] Re: 'apt-mark showauto' and 'apt show' is slow

2021-03-22 Thread Julian Andres Klode
Yes

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1713219

Title:
  'apt-mark showauto' and 'apt show' is slow

Status in apt package in Ubuntu:
  Triaged

Bug description:
  $ time apt-mark showauto >/dev/null

  real  0m0.587s
  user  0m0.552s
  sys   0m0.016s

  When I run the command first time, it is even much slower.

  I could do the job in fraction of a time using awk in POSIX shell
  script:

  auto_file='/var/lib/apt/extended_states'
  eval $(apt-config shell auto_file Dir::State::extended_states/f)
  awk '/^Package:/ {
pkg=$2
getline; arch=$2
getline
if($2==1) print pkg ":" arch
  }' "$auto_file" | CL_ALL=C sort -u

  real  0m0.019s
  user  0m0.008s
  sys   0m0.000s

  That prints architecture for every package and shows entries in
  slightly different order, though. And the file could be out of date
  showing packages that are not installed?!?

  Similarly

  apt show 

  is slow. (It also shows whether a package is manually or automatically
  installed.

  ProblemType: Bug
  DistroRelease: Ubuntu 14.04
  Package: apt 1.0.1ubuntu2.17
  ProcVersionSignature: Ubuntu 4.4.0-92.115~14.04.1-generic 4.4.76
  Uname: Linux 4.4.0-92-generic x86_64
  ApportVersion: 2.14.1-0ubuntu3.25
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sat Aug 26 12:59:00 2017
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2014-09-21 (1070 days ago)
  InstallationMedia: Ubuntu-Studio 14.04.1 LTS "Trusty Tahr" - Release amd64 
(20140722.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.cron.daily.apt: [modified]
  modified.conffile..etc.kernel.postinst.d.apt.auto.removal: [modified]
  mtime.conffile..etc.cron.daily.apt: 2017-05-03T10:27:27.617839
  mtime.conffile..etc.kernel.postinst.d.apt.auto.removal: 
2017-06-01T14:39:39.236080

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1713219/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1713219] Re: 'apt-mark showauto' and 'apt show' is slow

2021-03-22 Thread Julian Andres Klode
There is no bug there, this is all working correctly.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1713219

Title:
  'apt-mark showauto' and 'apt show' is slow

Status in apt package in Ubuntu:
  Triaged

Bug description:
  $ time apt-mark showauto >/dev/null

  real  0m0.587s
  user  0m0.552s
  sys   0m0.016s

  When I run the command first time, it is even much slower.

  I could do the job in fraction of a time using awk in POSIX shell
  script:

  auto_file='/var/lib/apt/extended_states'
  eval $(apt-config shell auto_file Dir::State::extended_states/f)
  awk '/^Package:/ {
pkg=$2
getline; arch=$2
getline
if($2==1) print pkg ":" arch
  }' "$auto_file" | CL_ALL=C sort -u

  real  0m0.019s
  user  0m0.008s
  sys   0m0.000s

  That prints architecture for every package and shows entries in
  slightly different order, though. And the file could be out of date
  showing packages that are not installed?!?

  Similarly

  apt show 

  is slow. (It also shows whether a package is manually or automatically
  installed.

  ProblemType: Bug
  DistroRelease: Ubuntu 14.04
  Package: apt 1.0.1ubuntu2.17
  ProcVersionSignature: Ubuntu 4.4.0-92.115~14.04.1-generic 4.4.76
  Uname: Linux 4.4.0-92-generic x86_64
  ApportVersion: 2.14.1-0ubuntu3.25
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sat Aug 26 12:59:00 2017
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2014-09-21 (1070 days ago)
  InstallationMedia: Ubuntu-Studio 14.04.1 LTS "Trusty Tahr" - Release amd64 
(20140722.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.cron.daily.apt: [modified]
  modified.conffile..etc.kernel.postinst.d.apt.auto.removal: [modified]
  mtime.conffile..etc.cron.daily.apt: 2017-05-03T10:27:27.617839
  mtime.conffile..etc.kernel.postinst.d.apt.auto.removal: 
2017-06-01T14:39:39.236080

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1713219/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1713219] Re: 'apt-mark showauto' and 'apt show' is slow

2021-03-22 Thread Julian Andres Klode
That's simply not true.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1713219

Title:
  'apt-mark showauto' and 'apt show' is slow

Status in apt package in Ubuntu:
  Triaged

Bug description:
  $ time apt-mark showauto >/dev/null

  real  0m0.587s
  user  0m0.552s
  sys   0m0.016s

  When I run the command first time, it is even much slower.

  I could do the job in fraction of a time using awk in POSIX shell
  script:

  auto_file='/var/lib/apt/extended_states'
  eval $(apt-config shell auto_file Dir::State::extended_states/f)
  awk '/^Package:/ {
pkg=$2
getline; arch=$2
getline
if($2==1) print pkg ":" arch
  }' "$auto_file" | CL_ALL=C sort -u

  real  0m0.019s
  user  0m0.008s
  sys   0m0.000s

  That prints architecture for every package and shows entries in
  slightly different order, though. And the file could be out of date
  showing packages that are not installed?!?

  Similarly

  apt show 

  is slow. (It also shows whether a package is manually or automatically
  installed.

  ProblemType: Bug
  DistroRelease: Ubuntu 14.04
  Package: apt 1.0.1ubuntu2.17
  ProcVersionSignature: Ubuntu 4.4.0-92.115~14.04.1-generic 4.4.76
  Uname: Linux 4.4.0-92-generic x86_64
  ApportVersion: 2.14.1-0ubuntu3.25
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sat Aug 26 12:59:00 2017
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2014-09-21 (1070 days ago)
  InstallationMedia: Ubuntu-Studio 14.04.1 LTS "Trusty Tahr" - Release amd64 
(20140722.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.cron.daily.apt: [modified]
  modified.conffile..etc.kernel.postinst.d.apt.auto.removal: [modified]
  mtime.conffile..etc.cron.daily.apt: 2017-05-03T10:27:27.617839
  mtime.conffile..etc.kernel.postinst.d.apt.auto.removal: 
2017-06-01T14:39:39.236080

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1713219/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1779767] Re: Default cron PATH does not include /snap/bin

2021-03-16 Thread Julian Andres Klode
** Changed in: cron (Ubuntu Hirsute)
   Status: Confirmed => Fix Committed

** Changed in: anacron (Ubuntu Hirsute)
   Status: Confirmed => Fix Committed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to cron in Ubuntu.
https://bugs.launchpad.net/bugs/1779767

Title:
  Default cron PATH does not include /snap/bin

Status in anacron package in Ubuntu:
  Fix Committed
Status in cron package in Ubuntu:
  Fix Committed
Status in anacron source package in Bionic:
  Confirmed
Status in cron source package in Bionic:
  Confirmed
Status in anacron source package in Focal:
  Confirmed
Status in cron source package in Focal:
  Confirmed
Status in anacron source package in Groovy:
  Confirmed
Status in cron source package in Groovy:
  Confirmed
Status in anacron source package in Hirsute:
  Fix Committed
Status in cron source package in Hirsute:
  Fix Committed

Bug description:
  I recently changed from a .deb install of LXD to a snap, and was
  surprised that one of my crontab scripts stopped working.

  I see that $PATH in a cron script only contains "/usr/bin:/bin",
  whereas my default shell also includes "/snap/bin".

  It seems to me that for the best user experience with snaps,
  "/snap/bin" should be part of the default $PATH in cron.

  ProblemType: Bug
  DistroRelease: Ubuntu 18.04
  Package: cron 3.0pl1-128.1ubuntu1
  ProcVersionSignature: Ubuntu 4.15.0-20.21-generic 4.15.17
  Uname: Linux 4.15.0-20-generic x86_64
  NonfreeKernelModules: kpatch_livepatch_Ubuntu_4_15_0_20_21_generic_40 
kpatch_livepatch_Ubuntu_4_15_0_20_21_generic_39 
livepatch_livepatch_Ubuntu_4_15_0_20_21_generic_ zfs zunicode zavl icp zcommon 
znvpair
  ApportVersion: 2.20.9-0ubuntu7.2
  Architecture: amd64
  CurrentDesktop: ubuntu:GNOME
  Date: Mon Jul  2 14:30:06 2018
  InstallationDate: Installed on 2017-12-20 (194 days ago)
  InstallationMedia: Ubuntu 18.04 LTS "Bionic Beaver" - Alpha amd64 (20171219)
  ProcEnviron:
   TERM=xterm-256color
   PATH=(custom, no user)
   XDG_RUNTIME_DIR=
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  SourcePackage: cron
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/anacron/+bug/1779767/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918928] [NEW] APT 2021/03 SRU release scheduling

2021-03-12 Thread Julian Andres Klode
Public bug reported:

We want to release the APT SRUs from Mar 2021 in a staggered manner,
such that we have 2-3 days between each release to get more chance to
discover regressions before rolling out to older releases.

Hence this bug, which we tag block-proposed-{bionic,focal} and then
untag once the delay has passed.

For groovy, we just want the normal aging :-)

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: In Progress


** Tags: block-proposed-bionic block-proposed-focal

** Changed in: apt (Ubuntu)
   Status: New => In Progress

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918928

Title:
  APT 2021/03 SRU release scheduling

Status in apt package in Ubuntu:
  In Progress

Bug description:
  We want to release the APT SRUs from Mar 2021 in a staggered manner,
  such that we have 2-3 days between each release to get more chance to
  discover regressions before rolling out to older releases.

  Hence this bug, which we tag block-proposed-{bionic,focal} and then
  untag once the delay has passed.

  For groovy, we just want the normal aging :-)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918928/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1829860] Re: APT unlocks in same order as it locks

2021-03-12 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Xenial)
   Status: New => Won't Fix

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1829860

Title:
  APT unlocks in same order as it locks

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Won't Fix
Status in apt source package in Bionic:
  Fix Released
Status in apt source package in Disco:
  Fix Released

Bug description:
  [Impact]
  APT releases the locks in the same order it acquires them, rather than 
reverse order. Given that we have no waiting for locks, this is not _super_ 
problematic, but it might be wrong: You'd get a lock failure on dpkg's lock, 
rather than lock-frontend.

  [Test case]
  Watch lock release with strace and see that it unlocks the right way.

  [Regression potential]
  Some other locking races or something?

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1829860/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918907] Re: Default Acquire::AllowReleaseInfoChange::Suite to "true"

2021-03-12 Thread Julian Andres Klode
Fixed since 2.1.10, which is in groovy already.

** No longer affects: apt (Ubuntu Groovy)

** Changed in: apt (Ubuntu)
   Status: New => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918907

Title:
  Default Acquire::AllowReleaseInfoChange::Suite to "true"

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New

Bug description:
  [Impact]
  APT 1.6-2.0 deny third-party repositories from changing their Suite value. 
This change was reverted in later apt releases, and will be reverted in 1.8 in 
Debian as well, so for consistency, we want to push that change to stable

  [Test plan]
  We added test cases to our extensive integration test suite run by 
autopkgtest.

  [Where problems could occur]
  It's hard to imagine anything here, we simply turn the flag for allowed from 
false to true. We even add tests for Suite changes :)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918907/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918920] Re: Harden test for no new acquires after transaction abort

2021-03-12 Thread Julian Andres Klode
This is fixed in 2.2.2 which will be synced ASAP, but this bug was
created after 2.2.2 and hence is not closed in the changelog.

** Changed in: apt (Ubuntu)
   Status: New => Fix Committed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918920

Title:
  Harden test for no new acquires after transaction abort

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Groovy:
  New

Bug description:
  [Impact]
  test-pdiff-usage is somewhat flaky, especially on Debian, this makes it less 
flaky. No end user impact as it's a test-only change.

  [Test plan]
  Running autopkgtest, which runs our extensive integration test suite which 
includes the changed test.

  [Where problems could occur]
  No end user regression potential on its own, but might slightly change 
regression potential for future pdiff changes:

  Test approach is slightly different now. It still catches that updates
  fail correctly, but tests more concretely that a transaction was
  aborted rather than that no worker received work (which was not
  guaranteed, the work could be scheduled before it was aborted).

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918920/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918920] [NEW] Harden test for no new acquires after transaction abort

2021-03-12 Thread Julian Andres Klode
Public bug reported:

[Impact]
test-pdiff-usage is somewhat flaky, especially on Debian, this makes it less 
flaky. No end user impact as it's a test-only change.

[Test plan]
Running autopkgtest, which runs our extensive integration test suite which 
includes the changed test.

[Where problems could occur]
No end user regression potential on its own, but might slightly change 
regression potential for future pdiff changes:

Test approach is slightly different now. It still catches that updates
fail correctly, but tests more concretely that a transaction was aborted
rather than that no worker received work (which was not guaranteed, the
work could be scheduled before it was aborted).

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: Fix Committed

** Affects: apt (Ubuntu Bionic)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Focal)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Groovy)
 Importance: Undecided
 Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Groovy)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918920

Title:
  Harden test for no new acquires after transaction abort

Status in apt package in Ubuntu:
  Fix Committed
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Groovy:
  New

Bug description:
  [Impact]
  test-pdiff-usage is somewhat flaky, especially on Debian, this makes it less 
flaky. No end user impact as it's a test-only change.

  [Test plan]
  Running autopkgtest, which runs our extensive integration test suite which 
includes the changed test.

  [Where problems could occur]
  No end user regression potential on its own, but might slightly change 
regression potential for future pdiff changes:

  Test approach is slightly different now. It still catches that updates
  fail correctly, but tests more concretely that a transaction was
  aborted rather than that no worker received work (which was not
  guaranteed, the work could be scheduled before it was aborted).

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918920/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918907] Re: Default Acquire::AllowReleaseInfoChange::Suite to "true"

2021-03-12 Thread Julian Andres Klode
** Also affects: apt (Ubuntu Bionic)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Groovy)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** No longer affects: apt (Ubuntu Hirsute)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918907

Title:
  Default Acquire::AllowReleaseInfoChange::Suite to "true"

Status in apt package in Ubuntu:
  New
Status in apt source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in apt source package in Groovy:
  New

Bug description:
  [Impact]
  APT 1.6-2.0 deny third-party repositories from changing their Suite value. 
This change was reverted in later apt releases, and will be reverted in 1.8 in 
Debian as well, so for consistency, we want to push that change to stable

  [Test plan]
  We added test cases to our extensive integration test suite run by 
autopkgtest.

  [Where problems could occur]
  It's hard to imagine anything here, we simply turn the flag for allowed from 
false to true. We even add tests for Suite changes :)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918907/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918907] [NEW] Default Acquire::AllowReleaseInfoChange::Suite to "true"

2021-03-12 Thread Julian Andres Klode
Public bug reported:

[Impact]
APT 1.6-2.0 deny third-party repositories from changing their Suite value. This 
change was reverted in later apt releases, and will be reverted in 1.8 in 
Debian as well, so for consistency, we want to push that change to stable

[Test plan]
We added test cases to our extensive integration test suite run by autopkgtest.

[Where problems could occur]
It's hard to imagine anything here, we simply turn the flag for allowed from 
false to true. We even add tests for Suite changes :)

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918907

Title:
  Default Acquire::AllowReleaseInfoChange::Suite to "true"

Status in apt package in Ubuntu:
  New

Bug description:
  [Impact]
  APT 1.6-2.0 deny third-party repositories from changing their Suite value. 
This change was reverted in later apt releases, and will be reverted in 1.8 in 
Debian as well, so for consistency, we want to push that change to stable

  [Test plan]
  We added test cases to our extensive integration test suite run by 
autopkgtest.

  [Where problems could occur]
  It's hard to imagine anything here, we simply turn the flag for allowed from 
false to true. We even add tests for Suite changes :)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918907/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1871268] Re: Installation fails due to useless immediate configuration error when "Install Third-Party Drivers" is selected

2021-03-10 Thread Julian Andres Klode
Corrected the task for bionic as per policy (in progress means SRU
uploaded).

I'll be working on the bionic SRU tomorrow, I've just prepared SRUs for
groovy and focal for other issues, and will be cherry-picking these
patches on top of those :)

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1871268

Title:
  Installation fails due to useless immediate configuration error when
  "Install Third-Party Drivers" is selected

Status in Ubuntu CD Images:
  Fix Released
Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Triaged
Status in apt source package in Focal:
  Fix Released
Status in apt source package in Groovy:
  Fix Released
Status in apt package in Debian:
  Fix Released

Bug description:
  [Impact]
  Installations that really succeeded would then fail because APT could not 
immediately configure a package. Which is a pointless way to fail at that 
point, because everything did work out anyway.

  We have two changes that help address this:

  * The first one stops immediately configuring multi-arch siblings
  (e.g. libc6:i386 when it's configuring libc6:amd64). This was not
  necessary, and caused all the libc6:i386 failures here.

  * The second change sort of also supersedes the first one: It just
  ignores any errors from immediate configuration, relying on the fact
  that it's checked and rectified at a later point if there are
  unconfigured packages (which is what made all those failures happen
  spuriously after having successfully installed everything).

  [Test case]
  We have one test case in EIPP format in the Debian bug 973305 which was only 
helped by the second change, not the first one. Run /usr/lib/apt/planners < 
eipp.log and check there are no errors.

  TODO: It's unclear if the APT from proposed installed in the live
  session will fix the installer, needs investigation, but would make a
  useful test case.

  [Regression potential]
  It's imaginable that we missed something somewhere and some path that checked 
for a set error doesn't check it anymore, and we report success when we hit an 
error, but it seems unlikely.

  Behavior of --simulate changes. This used to fail before as well, and
  will now only produce a warning. We don't believe that is a reason of
  concern.

  [Groovy SRU]
  The groovy SRU is a sync of the 2.1.11 micro release from Debian unstable 
which also incorporates changes to the documentation: A typo fix, replacing 
focal with groovy in examples, and minor Dutch manual pages translation updates.

  We do not have test cases for the documentation changes, and we do not
  consider there to be a huge regression potential. As long as they
  build, they should be readable - maybe some words are wrong in the
  translation, who knows.

  [Original bug report]
  Test Case
  1. Install Ubuntu Desktop on hardware with an nVidia card and select to 
install 3rd party drivers
  2. Proceed with installation

  The following error message is displayed in /var/log/syslog
  /plugininstall.py: Verifying downloads ...
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/g/gcc-defaults/gcc_9.3.0-1ubuntu2_amd64.deb: "Version: 
'9.3.0-1ubuntu2' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxcrypt/libcrypt-dev_4.4.10-10ubuntu4_amd64.deb: 
"Version: '4.4.10-10ubuntu4' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/g/gcc-defaults/g++_9.3.0-1ubuntu2_amd64.deb: "Version: 
'9.3.0-1ubuntu2' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/z/zlib/zlib1g_1.2.11.dfsg-2ubuntu1_i386.deb: "Version: 
'1.2.11.dfsg-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxau/libxau6_1.0.9-0ubuntu1_i386.deb: "Version: 
'1.0.9-0ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxdmcp/libxdmcp6_1.1.3-0ubuntu1_i386.deb: "Version: 
'1.1.3-0ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libx11/libx11-6_1.6.9-2ubuntu1_i386.deb: "Version: 
'1.6.9-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxext/libxext6_1.3.4-0ubuntu1_i386.deb: "Version: 
'1.3.4-0ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/l/lm-sensors/libsensors5_3.6.0-2ubuntu1_i386.deb: "Version: 
'3.6.0-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libx11/libx11-xcb1_1.6.9-2ubuntu1_i386.deb: "Version: 
'1.6.9-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxdamage/libxdamage1_1.1.5-1_i386.deb: "Version: 
'1.1.5-1' not found."
  /plugininstall.py: Failed to find package object for 

[Touch-packages] [Bug 1871268] Re: Installation fails due to useless immediate configuration error when "Install Third-Party Drivers" is selected

2021-03-10 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Bionic)
   Status: In Progress => Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1871268

Title:
  Installation fails due to useless immediate configuration error when
  "Install Third-Party Drivers" is selected

Status in Ubuntu CD Images:
  Fix Released
Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Triaged
Status in apt source package in Focal:
  Fix Released
Status in apt source package in Groovy:
  Fix Released
Status in apt package in Debian:
  Fix Released

Bug description:
  [Impact]
  Installations that really succeeded would then fail because APT could not 
immediately configure a package. Which is a pointless way to fail at that 
point, because everything did work out anyway.

  We have two changes that help address this:

  * The first one stops immediately configuring multi-arch siblings
  (e.g. libc6:i386 when it's configuring libc6:amd64). This was not
  necessary, and caused all the libc6:i386 failures here.

  * The second change sort of also supersedes the first one: It just
  ignores any errors from immediate configuration, relying on the fact
  that it's checked and rectified at a later point if there are
  unconfigured packages (which is what made all those failures happen
  spuriously after having successfully installed everything).

  [Test case]
  We have one test case in EIPP format in the Debian bug 973305 which was only 
helped by the second change, not the first one. Run /usr/lib/apt/planners < 
eipp.log and check there are no errors.

  TODO: It's unclear if the APT from proposed installed in the live
  session will fix the installer, needs investigation, but would make a
  useful test case.

  [Regression potential]
  It's imaginable that we missed something somewhere and some path that checked 
for a set error doesn't check it anymore, and we report success when we hit an 
error, but it seems unlikely.

  Behavior of --simulate changes. This used to fail before as well, and
  will now only produce a warning. We don't believe that is a reason of
  concern.

  [Groovy SRU]
  The groovy SRU is a sync of the 2.1.11 micro release from Debian unstable 
which also incorporates changes to the documentation: A typo fix, replacing 
focal with groovy in examples, and minor Dutch manual pages translation updates.

  We do not have test cases for the documentation changes, and we do not
  consider there to be a huge regression potential. As long as they
  build, they should be readable - maybe some words are wrong in the
  translation, who knows.

  [Original bug report]
  Test Case
  1. Install Ubuntu Desktop on hardware with an nVidia card and select to 
install 3rd party drivers
  2. Proceed with installation

  The following error message is displayed in /var/log/syslog
  /plugininstall.py: Verifying downloads ...
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/g/gcc-defaults/gcc_9.3.0-1ubuntu2_amd64.deb: "Version: 
'9.3.0-1ubuntu2' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxcrypt/libcrypt-dev_4.4.10-10ubuntu4_amd64.deb: 
"Version: '4.4.10-10ubuntu4' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/g/gcc-defaults/g++_9.3.0-1ubuntu2_amd64.deb: "Version: 
'9.3.0-1ubuntu2' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/z/zlib/zlib1g_1.2.11.dfsg-2ubuntu1_i386.deb: "Version: 
'1.2.11.dfsg-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxau/libxau6_1.0.9-0ubuntu1_i386.deb: "Version: 
'1.0.9-0ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxdmcp/libxdmcp6_1.1.3-0ubuntu1_i386.deb: "Version: 
'1.1.3-0ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libx11/libx11-6_1.6.9-2ubuntu1_i386.deb: "Version: 
'1.6.9-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxext/libxext6_1.3.4-0ubuntu1_i386.deb: "Version: 
'1.3.4-0ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/l/lm-sensors/libsensors5_3.6.0-2ubuntu1_i386.deb: "Version: 
'3.6.0-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libx11/libx11-xcb1_1.6.9-2ubuntu1_i386.deb: "Version: 
'1.6.9-2ubuntu1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxdamage/libxdamage1_1.1.5-1_i386.deb: "Version: 
'1.1.5-1' not found."
  /plugininstall.py: Failed to find package object for 
/cdrom//pool/main/libx/libxfixes/libxfixes3_5.0.3-1_i386.deb: "Version: 
'5.0.3-1' not found."
  /plugininstall.py: Failed to find package object for 

[Touch-packages] [Bug 1898026] Re: interruption of dist-upgrade can leave you next release in sources.list

2021-03-10 Thread Julian Andres Klode
For the record, we are going to roll the apt change out in groovy
shortly, but delay the LTS updates to give it some more real world
testing.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1898026

Title:
  interruption of dist-upgrade can leave you next release in
  sources.list

Status in apt package in Ubuntu:
  Fix Released
Status in ubuntu-release-upgrader package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  New
Status in ubuntu-release-upgrader source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in ubuntu-release-upgrader source package in Focal:
  New
Status in apt source package in Groovy:
  In Progress
Status in ubuntu-release-upgrader source package in Groovy:
  New
Status in apt source package in Hirsute:
  Fix Released
Status in ubuntu-release-upgrader source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  Calling update() or installing packages from apt clients resets their SIGINT 
and SIGQUIT handlers to SIG_DFL, overriding any signal handlers they might have 
set for them.

  In case of ubuntu-release-upgrader, this results in the release
  upgrader being unable to handle interrupts after it did the initial
  update - the default libc handler will run and the program exits.

  [Test plan]

  [[apt]]
  As a standalone test for apt, we can test the following script:

  import apt
  import time

  apt.Cache().update()
  print("WAITING")
  try:
  time.sleep(1)
  except BaseException as e:
  print("Seen", repr(e))
  print("END")

  Pressing Ctrl+C while WAITING is printed should print Seen
  KeyboardInterrupt, and importantly, also the END line.

  [Where problems could occur]
  apt: This specific change removes the two lines that SIG_DFL the signal 
handlers after running scripts. AFAWCT those lines are unnecessary - the code 
that calls it temporarily sets the handlers to SIG_IGN but restores previous 
handlers at the end; it was wrongly refactored decades ago. A regression could 
occur in that those signals will now continue to be ignored if we missed a spot.

  [Original bug report]
  As a long-time-user of ?ubuntu, with apt-get as tool of choice for 
updates/upgrades I wrote a daily script for updates, with 'dist-upgrade'.
  In all earlier years, it wouldn't actually do an upgrade of a ?ubuntu 
version; just all packages including new ones. Version updates had to be 
initiated manually, and I was always asked if I really wanted the new ?ubuntu 
version. Sounds appropriate.

  Last night when it (dist-upgrade), it just gave me 20.04. No questions asked. 
I for one consider this kind of intrusive, though.
  It *might* have to make with me trying 'sudo do-release-upgrade -m desktop' a 
number of times earlier; just to *check* if the upgrade was on offer; but this 
is only a guess.

  In *any* case, a pop-up asking "Are you sure? Are you connected
  through an adequate pipe? Are you sitting with full batteries; better
  a power supply?" would be convenient; since I am using my PC for
  urgent duties, and didn't want to fiddle with unexpected upgrade bugs.

  ProblemType: Bug
  DistroRelease: Ubuntu 20.04
  Package: ubuntu-release-upgrader-core 1:20.04.25
  ProcVersionSignature: Ubuntu 5.4.0-48.52-generic 5.4.60
  Uname: Linux 5.4.0-48-generic x86_64
  ApportVersion: 2.20.11-0ubuntu27.9
  Architecture: amd64
  CasperMD5CheckResult: skip
  CrashDB: ubuntu
  CurrentDesktop: KDE
  Date: Thu Oct  1 11:48:39 2020
  InstallationDate: Installed on 2019-03-14 (566 days ago)
  InstallationMedia: Kubuntu 18.04.1 LTS "Bionic Beaver" - Release amd64 
(20180725)
  PackageArchitecture: all
  SourcePackage: ubuntu-release-upgrader
  Symptom: ubuntu-release-upgrader
  UpgradeStatus: Upgraded to focal on 2020-09-29 (1 days ago)
  VarLogDistupgradeAptlog:
   Log time: 2020-09-29 19:44:28.696289
   Log time: 2020-09-29 19:44:33.573481

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1898026/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1916725] Re: Protected/Important packages are not deconfigured, require Force-LoopBreak

2021-03-10 Thread Julian Andres Klode
** Description changed:

  [Impact]
  If a package that is Protected: yes (or Important: yes), or one of it's 
dependencies, is involved in a dependency loop with Breaks, APT requires 
APT::Force-LoopBreak instead of resolving the situation directly.
+ 
+ On focal, we also introduce the actual support for protected packages to
+ enable upgrading to later releases more easily (in case a protected
+ package needs to be removed during the upgrade), and to make the
+ backport more similar to main.
  
  [Test plan]
  Run the integration test suite (the autopkgtest) :)
  
  Our test suite covers the tests for both Breaks and Conflicts.
  
  Breaks:
  
  protected-sysvinit (= 1) without dependencies is installed
  protected-sysvinit (= 2) Pre-Depends protected-systemd-sysv
  protected-systemd-sysv (= 2) Breaks: protected-sysvinit (<< 2)
  
  Test: Install protected-sysvinit (= 2)
  Expected result: Unpacking protected-sysvinit (= 2) deconfigures 
protected-sysvinit (= 1), and then we unpack and configure protected-sysvinit 
(= 2) and end up with a working system.
  
  Conflicts: As for Breaks, but the Conflicts will remove the package
  temporarily, requiring the use of APT::Force-LoopBreak option.
  
+ For focal, we also do have a test to check that the Protected field is
+ being used.
+ 
  [Where problems could occur]
  We now allow dpkg to automatically deconfigure protected packages. This 
should just make them behave like normal packages to APT's eye, but bugs I 
guess could occur somewhere in the APT/dpkg interaction (this only applies to 
releases with Protected support in dpkg, Important is not affected, it's always 
been "normal" for dpkg).
  
  During development, we accidentally simplified the patch so much that
  Conflicts did not require Force-LoopBreak for temporary removal. We
  fixed that, but it points out that there is a place where the loop break
  check happens that is a potential regression place.
+ 
+ On focal, we pass additional flags to dpkg that focal's dpkg does not
+ understand, however, we only do that if dpkg asserts it does that, so in
+ practice, this should all work fine and the code path will only be taken
+ with >=groovy dpkg.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1916725

Title:
  Protected/Important packages are not deconfigured, require Force-
  LoopBreak

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Bionic:
  Triaged
Status in apt source package in Focal:
  Triaged
Status in apt source package in Groovy:
  In Progress
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  If a package that is Protected: yes (or Important: yes), or one of it's 
dependencies, is involved in a dependency loop with Breaks, APT requires 
APT::Force-LoopBreak instead of resolving the situation directly.

  On focal, we also introduce the actual support for protected packages
  to enable upgrading to later releases more easily (in case a protected
  package needs to be removed during the upgrade), and to make the
  backport more similar to main.

  [Test plan]
  Run the integration test suite (the autopkgtest) :)

  Our test suite covers the tests for both Breaks and Conflicts.

  Breaks:

  protected-sysvinit (= 1) without dependencies is installed
  protected-sysvinit (= 2) Pre-Depends protected-systemd-sysv
  protected-systemd-sysv (= 2) Breaks: protected-sysvinit (<< 2)

  Test: Install protected-sysvinit (= 2)
  Expected result: Unpacking protected-sysvinit (= 2) deconfigures 
protected-sysvinit (= 1), and then we unpack and configure protected-sysvinit 
(= 2) and end up with a working system.

  Conflicts: As for Breaks, but the Conflicts will remove the package
  temporarily, requiring the use of APT::Force-LoopBreak option.

  For focal, we also do have a test to check that the Protected field is
  being used.

  [Where problems could occur]
  We now allow dpkg to automatically deconfigure protected packages. This 
should just make them behave like normal packages to APT's eye, but bugs I 
guess could occur somewhere in the APT/dpkg interaction (this only applies to 
releases with Protected support in dpkg, Important is not affected, it's always 
been "normal" for dpkg).

  During development, we accidentally simplified the patch so much that
  Conflicts did not require Force-LoopBreak for temporary removal. We
  fixed that, but it points out that there is a place where the loop
  break check happens that is a potential regression place.

  On focal, we pass additional flags to dpkg that focal's dpkg does not
  understand, however, we only do that if dpkg asserts it does that, so
  in practice, this should all work fine and the code path will only be
  taken with >=groovy dpkg.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1916725/+subscriptions


[Touch-packages] [Bug 1918112] Re: Deal with rred shortcomings around empty patch files

2021-03-08 Thread Julian Andres Klode
We're actually not backporting this to stable releases just future
stable one :)

** Changed in: apt (Ubuntu Groovy)
   Status: New => Won't Fix

** Changed in: apt (Ubuntu Focal)
   Status: New => Won't Fix

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918112

Title:
  Deal with rred shortcomings around empty patch files

Status in apt package in Ubuntu:
  New
Status in apt source package in Focal:
  Won't Fix
Status in apt source package in Groovy:
  Won't Fix
Status in apt source package in Hirsute:
  New

Bug description:
  [Impact]
  Merge request https://salsa.debian.org/apt-team/apt/-/merge_requests/159 has 
a bunch of fixes for empty pdiff files, cleaning up pdiff files, as well as 
correcting some bugs with index files in the mirror method.

  While we don't use pdiffs ourselves, third party repositories might,
  and people might want to use rred on Ubuntu to create merged pdiff
  files, and hence benefit from those patches, so we want to backport it
  to 2.0, such that it is not more buggy than Debian's 1.8 and forces
  you to run Debian if you want to do those things.

  [Test plan]
  Each commit comes with a change to the test suite that tests it, and the test 
suite is fairly exhaustive too.

  [Where problems could occur]
  Changes are limited to adding headers to Index file downloads (so that mirror 
works correctly), and pdiff-specific code, so problems could occur with mirror 
method or pdiff.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918112/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1918112] [NEW] Deal with rred shortcomings around empty patch files

2021-03-08 Thread Julian Andres Klode
Public bug reported:

[Impact]
Merge request https://salsa.debian.org/apt-team/apt/-/merge_requests/159 has a 
bunch of fixes for empty pdiff files, cleaning up pdiff files, as well as 
correcting some bugs with index files in the mirror method.

While we don't use pdiffs ourselves, third party repositories might, and
people might want to use rred on Ubuntu to create merged pdiff files,
and hence benefit from those patches, so we want to backport it to 2.0,
such that it is not more buggy than Debian's 1.8 and forces you to run
Debian if you want to do those things.

[Test plan]
Each commit comes with a change to the test suite that tests it, and the test 
suite is fairly exhaustive too.

[Where problems could occur]
Changes are limited to adding headers to Index file downloads (so that mirror 
works correctly), and pdiff-specific code, so problems could occur with mirror 
method or pdiff.

** Affects: apt (Ubuntu)
 Importance: Undecided
 Status: New

** Affects: apt (Ubuntu Focal)
 Importance: Undecided
 Status: Won't Fix

** Affects: apt (Ubuntu Groovy)
 Importance: Undecided
 Status: Won't Fix

** Affects: apt (Ubuntu Hirsute)
 Importance: Undecided
 Status: New

** Also affects: apt (Ubuntu Focal)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Hirsute)
   Importance: Undecided
   Status: New

** Also affects: apt (Ubuntu Groovy)
   Importance: Undecided
   Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1918112

Title:
  Deal with rred shortcomings around empty patch files

Status in apt package in Ubuntu:
  New
Status in apt source package in Focal:
  Won't Fix
Status in apt source package in Groovy:
  Won't Fix
Status in apt source package in Hirsute:
  New

Bug description:
  [Impact]
  Merge request https://salsa.debian.org/apt-team/apt/-/merge_requests/159 has 
a bunch of fixes for empty pdiff files, cleaning up pdiff files, as well as 
correcting some bugs with index files in the mirror method.

  While we don't use pdiffs ourselves, third party repositories might,
  and people might want to use rred on Ubuntu to create merged pdiff
  files, and hence benefit from those patches, so we want to backport it
  to 2.0, such that it is not more buggy than Debian's 1.8 and forces
  you to run Debian if you want to do those things.

  [Test plan]
  Each commit comes with a change to the test suite that tests it, and the test 
suite is fairly exhaustive too.

  [Where problems could occur]
  Changes are limited to adding headers to Index file downloads (so that mirror 
works correctly), and pdiff-specific code, so problems could occur with mirror 
method or pdiff.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1918112/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1415701] Re: apt is unable to handle CDROMS with default mount option

2021-03-05 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Xenial)
   Status: Triaged => Won't Fix

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1415701

Title:
  apt is unable to handle CDROMS with default mount option

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Won't Fix

Bug description:
  I have found i am almost completely unable to use CDROM as an apt
  repository on Ubuntu 14.04

  
  I found the command apt-cdrom add by default, was unable to detect CDROMs
  it was attempting to find them in /media/cdrom/
  by default ubuntu now seems to mount cds in /media/$USER/ and this is not 
something set via fstab 
  i was able to get them to add with the command 
  apt-cdrom add -m --cdrom /media/$USER/

  I seem to have been able to pull the package list from the CD, but when it 
comes to installing from it, apt encounters a similar error
  it ask you to insert the disc in drive /media/cdrom/ again


  I also am completely unable to add cds via the software & updates
  setting manager, it always fails to find the cd, which is presumably
  part of the same issue.

  
  manually remounting with 
  sudo mount /dev/cdrom /media/cdrom 
  get some things to work, but most of the actions trigger a remount which goes 
back to the default location


  I was finally able to get things working again properly by adding the fstab  
line 
  /dev/cdrom  /media/cdrom  udf,iso9660  user,noauto,exec,utf8  0  0
  resolves the problem for me
  but this is a work around not a fix, it may break other stuff

  
  In apt conf i have tried setting 
  Acquire::cdrom::mount "file/path/";
  with various file paths, the file paths, to try and make it work with the 
default mount behaviour,this looked to be the right setting to change as the 
message changed accordingly to what i was setting, but i could not get it to 
work  for actually installing packages, even by setting an absolute file path 
for my user in the media directory.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1415701/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1589204] Re: apt-get dist-upgrade and apt full-upgrade not reporting held back package

2021-03-05 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Xenial)
   Status: Confirmed => Won't Fix

** Changed in: apt (Ubuntu Trusty)
   Status: Confirmed => Won't Fix

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1589204

Title:
  apt-get dist-upgrade and apt full-upgrade not reporting held back
  package

Status in apt package in Ubuntu:
  Confirmed
Status in apt source package in Trusty:
  Won't Fix
Status in apt source package in Xenial:
  Won't Fix

Bug description:
  I have the situation where a package cannot be upgraded because of missing 
dependencies.  However
  sudo apt-get dist-upgrade
  and
  sudo apt full-upgrade
  and
  sudo apt upgrade
  all show
  0 to upgrade, 0 to newly install, 0 to remove and 0 not to upgrade.

  Only
  sudo apt-get upgrade
  shows
  0 to upgrade, 0 to newly install, 0 to remove and 1 not to upgrade.

  I believe all of the above commands should show that 1 cannot be
  upgraded.

  This bug is not about the particular package I am trying to install, but 
about the way apt reports the problem.
  The details, for reference, are that I am trying to upgrade mosquitto from 
the mosquitto ppa.  After apt-get update I see
  $ apt-cache policy mosquitto
  mosquitto:
    Installed: 1.4.8-1build1
    Candidate: 1.4.9-0mosquitto1
    Version table:
   1.4.9-0mosquitto1 500
  500 http://ppa.launchpad.net/mosquitto-dev/mosquitto-ppa/ubuntu 
xenial/main amd64 Packages
   *** 1.4.8-1build1 500
  500 http://gb.archive.ubuntu.com/ubuntu xenial/universe amd64 Packages
  100 /var/lib/dpkg/status

  but if I try to manually install mosquitto I get

  The following packages have unmet dependencies.
   mosquitto : Depends: libwebsockets3 (>= 1.2) but it is not installable

  and

  $ apt-cache policy libwebsockets3
  libwebsockets3:
    Installed: (none)
    Candidate: (none)
  Due to the fact that libwebsockets3 is not available for Xenial.
  As I said this bug is not about the problem with this particular package but 
about the fact that apt does not report that there is a package to be upgraded, 
but it cannot be done.

  ProblemType: Bug
  DistroRelease: Ubuntu 16.04
  Package: apt 1.2.12~ubuntu16.04.1
  ProcVersionSignature: Ubuntu 4.4.0-22.40-generic 4.4.8
  Uname: Linux 4.4.0-22-generic x86_64
  ApportVersion: 2.20.1-0ubuntu2.1
  Architecture: amd64
  CurrentDesktop: Unity
  Date: Sun Jun  5 08:45:33 2016
  InstallationDate: Installed on 2014-10-21 (592 days ago)
  InstallationMedia: Ubuntu 14.10 "Utopic Unicorn" - Alpha amd64 (20141017)
  SourcePackage: apt
  UpgradeStatus: Upgraded to xenial on 2016-03-06 (90 days ago)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1589204/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1615381] Re: apt-get autoremove may remove current kernel

2021-03-05 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Groovy)
   Status: Triaged => In Progress

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to unattended-upgrades in
Ubuntu.
https://bugs.launchpad.net/bugs/1615381

Title:
  apt-get autoremove may remove current kernel

Status in apt package in Ubuntu:
  Fix Released
Status in unattended-upgrades package in Ubuntu:
  Fix Released
Status in unattended-upgrades source package in Trusty:
  Won't Fix
Status in unattended-upgrades source package in Xenial:
  Fix Released
Status in unattended-upgrades source package in Artful:
  Won't Fix
Status in apt source package in Bionic:
  Confirmed
Status in apt source package in Focal:
  Confirmed
Status in apt source package in Groovy:
  In Progress
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  APT will try and fail to remove the currently running kernel, when booted 
into an older kernel that is not protected. May wreak some havoc if mixed with 
other operations to the point that apt goes weird and needs manual fixing up.

  [Test case]
  For the SRUs we have included an automated test case that starts with an 
empty autoremove config file, and then checks that the running kernel is 
protected at runtime.

  You can also test manually, but it's not necessary:

  - Install new kernel
  - Reboot into kernel not listed as protected in 01autoremove-kernels
  - Run autoremove

  [Where problems could occur]
  We may more easily run out of space in /boot. hirsute has new autoremoval 
code that runs completely at runtime; but that seems a bit large to SRU after 
only a few weeks in hirsute. Hence, we should protect the current kernel _in 
addition_ to the other kernels, just like unattended-upgrades and 
update-manager do. This increases the risk of filling up /boot compared to 
older apt versions, but is at the same level as unattended-upgrades and 
update-manager.

  [Original bug report]

  This may happen, if you boot one of the older kernels, that is not
  protected by /etc/apt/apt.conf.d/01autoremove-kernels

  Workaround: run
  /etc/kernel/postinst.d/apt-auto-removal
  during each boot (e.g. by using cron).
  Note: The workaround breaks autoremoving feature of new unneeded kernels in  
unattended-upgrades i.e. the setting 
'Unattended-Upgrade::Remove-New-Unused-Dependencies "true"' (which is default 
in 16.04 unless 'Unattended-Upgrade::Remove-Unused-Dependencies "true"' is set 
in '/etc/apt/apt.conf.d/50unattended-upgrades'.

  In shell:

  $ uname -r
  4.4.0-22-generic
  $ apt-get -s autoremove
  NOTE: This is only a simulation!
    apt-get needs root privileges for real execution.
    Keep also in mind that locking is deactivated,
    so don't depend on the relevance to the real current situation!
  Reading package lists... Done
  Building dependency tree
  Reading state information... Done
  The following packages will be REMOVED:
    linux-headers-4.4.0-21 linux-headers-4.4.0-21-generic linux-headers-4.4.0-22
    linux-headers-4.4.0-22-generic linux-headers-4.4.0-31-generic
    linux-image-4.4.0-21-generic linux-image-4.4.0-22-generic
    linux-image-4.4.0-31-generic linux-image-extra-4.4.0-21-generic
    linux-image-extra-4.4.0-22-generic linux-image-extra-4.4.0-31-generic
  0 upgraded, 0 newly installed, 11 to remove and 13 not upgraded.

  ProblemType: Bug
  DistroRelease: Ubuntu 16.04
  Package: apt 1.2.12~ubuntu16.04.1
  ProcVersionSignature: Ubuntu 4.4.0-22.40-generic 4.4.8
  Uname: Linux 4.4.0-22-generic x86_64
  ApportVersion: 2.20.1-0ubuntu2.1
  Architecture: amd64
  CurrentDesktop: XFCE
  Date: Sun Aug 21 16:11:27 2016
  EcryptfsInUse: Yes
  InstallationDate: Installed on 2016-04-28 (114 days ago)
  InstallationMedia: Xubuntu 16.04 LTS "Xenial Xerus" - Release amd64 
(20160420.1)
  SourcePackage: apt
  UpgradeStatus: No upgrade log present (probably fresh install)
  modified.conffile..etc.kernel.postinst.d.apt-auto-removal: [modified]
  mtime.conffile..etc.kernel.postinst.d.apt-auto-removal: 
2016-07-30T12:15:32.706300

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1615381/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1898026] Re: interruption of dist-upgrade can leave you next release in sources.list

2021-03-05 Thread Julian Andres Klode
** Description changed:

- As a long-time-user of ?ubuntu, with apt-get as tool of choice for 
updates/upgrades I wrote a daily script for updates, with 'dist-upgrade'. 
+ [Impact]
+ Calling update() or installing packages from apt clients resets their SIGINT 
and SIGQUIT handlers to SIG_DFL, overriding any signal handlers they might have 
set for them.
+ 
+ In case of ubuntu-release-upgrader, this results in the release upgrader
+ being unable to handle interrupts after it did the initial update - the
+ default libc handler will run and the program exits.
+ 
+ [Test plan]
+ 
+ [[apt]]
+ As a standalone test for apt, we can test the following script:
+ 
+ import apt
+ import time
+ 
+ apt.Cache().update()
+ print("WAITING")
+ try:
+ time.sleep(1)
+ except BaseException as e:
+ print("Seen", repr(e))
+ print("END")
+ 
+ Pressing Ctrl+C while WAITING is printed should print Seen
+ KeyboardInterrupt, and importantly, also the END line.
+ 
+ [Where problems could occur]
+ apt: This specific change removes the two lines that SIG_DFL the signal 
handlers after running scripts. AFAWCT those lines are unnecessary - the code 
that calls it temporarily sets the handlers to SIG_IGN but restores previous 
handlers at the end; it was wrongly refactored decades ago. A regression could 
occur in that those signals will now continue to be ignored if we missed a spot.
+ 
+ [Original bug report]
+ As a long-time-user of ?ubuntu, with apt-get as tool of choice for 
updates/upgrades I wrote a daily script for updates, with 'dist-upgrade'.
  In all earlier years, it wouldn't actually do an upgrade of a ?ubuntu 
version; just all packages including new ones. Version updates had to be 
initiated manually, and I was always asked if I really wanted the new ?ubuntu 
version. Sounds appropriate.
  
- Last night when it (dist-upgrade), it just gave me 20.04. No questions asked. 
I for one consider this kind of intrusive, though. 
+ Last night when it (dist-upgrade), it just gave me 20.04. No questions asked. 
I for one consider this kind of intrusive, though.
  It *might* have to make with me trying 'sudo do-release-upgrade -m desktop' a 
number of times earlier; just to *check* if the upgrade was on offer; but this 
is only a guess.
  
  In *any* case, a pop-up asking "Are you sure? Are you connected through
  an adequate pipe? Are you sitting with full batteries; better a power
  supply?" would be convenient; since I am using my PC for urgent duties,
  and didn't want to fiddle with unexpected upgrade bugs.
  
  ProblemType: Bug
  DistroRelease: Ubuntu 20.04
  Package: ubuntu-release-upgrader-core 1:20.04.25
  ProcVersionSignature: Ubuntu 5.4.0-48.52-generic 5.4.60
  Uname: Linux 5.4.0-48-generic x86_64
  ApportVersion: 2.20.11-0ubuntu27.9
  Architecture: amd64
  CasperMD5CheckResult: skip
  CrashDB: ubuntu
  CurrentDesktop: KDE
  Date: Thu Oct  1 11:48:39 2020
  InstallationDate: Installed on 2019-03-14 (566 days ago)
  InstallationMedia: Kubuntu 18.04.1 LTS "Bionic Beaver" - Release amd64 
(20180725)
  PackageArchitecture: all
  SourcePackage: ubuntu-release-upgrader
  Symptom: ubuntu-release-upgrader
  UpgradeStatus: Upgraded to focal on 2020-09-29 (1 days ago)
  VarLogDistupgradeAptlog:
-  Log time: 2020-09-29 19:44:28.696289
-  Log time: 2020-09-29 19:44:33.573481
+  Log time: 2020-09-29 19:44:28.696289
+  Log time: 2020-09-29 19:44:33.573481

** Changed in: apt (Ubuntu Groovy)
   Status: Triaged => In Progress

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1898026

Title:
  interruption of dist-upgrade can leave you next release in
  sources.list

Status in apt package in Ubuntu:
  Fix Released
Status in ubuntu-release-upgrader package in Ubuntu:
  In Progress
Status in apt source package in Bionic:
  New
Status in ubuntu-release-upgrader source package in Bionic:
  New
Status in apt source package in Focal:
  New
Status in ubuntu-release-upgrader source package in Focal:
  New
Status in apt source package in Groovy:
  In Progress
Status in ubuntu-release-upgrader source package in Groovy:
  New
Status in apt source package in Hirsute:
  Fix Released
Status in ubuntu-release-upgrader source package in Hirsute:
  In Progress

Bug description:
  [Impact]
  Calling update() or installing packages from apt clients resets their SIGINT 
and SIGQUIT handlers to SIG_DFL, overriding any signal handlers they might have 
set for them.

  In case of ubuntu-release-upgrader, this results in the release
  upgrader being unable to handle interrupts after it did the initial
  update - the default libc handler will run and the program exits.

  [Test plan]

  [[apt]]
  As a standalone test for apt, we can test the following script:

  import apt
  import time

  apt.Cache().update()
  print("WAITING")
  try:
  time.sleep(1)
  except BaseException as e:
  print("Seen", repr(e))
  print("END")

  

[Touch-packages] [Bug 1693900] Re: apt-get update should return exit code != 0 on error

2021-03-05 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Groovy)
   Status: Triaged => In Progress

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1693900

Title:
  apt-get update should return exit code != 0 on error

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  New
Status in apt source package in Bionic:
  Triaged
Status in apt source package in Focal:
  Triaged
Status in apt source package in Groovy:
  In Progress

Bug description:
  [Impact]
  There is no way for scripts to run apt update and be sure that it was 
entirely successful.

  We introduce a new flag, --error-mode=any that makes apt update also
  error out on transient errors.

  [Test plan]
  We have included a test in the test-suite in 
test-apt-update-failure-propagation that ensures that warnings become errors.

  [Where problems could occur]
  If there's an issue some warnings for transient errors could become errors, 
or worse I suppose it's possible but unlikely for errors to be reported as 
warnings...

  Since the new behavior is behind a flag, regression potential for
  existing scripts is low if there is no bug.

  [Original bug report]

  When running 'apt-get update' (e.g. on a container install post-
  install script), apt-get return with exit code 0, even so it wasn't
  able to "update" properly. E.g.:

  + apt-get update
  Err:1 http://de.archive.ubuntu.com/ubuntu xenial InRelease
    Temporary failure resolving 'de.archive.ubuntu.com'
  Err:2 http://security.ubuntu.com/ubuntu xenial-security InRelease
    Temporary failure resolving 'security.ubuntu.com'
  Err:3 http://de.archive.ubuntu.com/ubuntu xenial-updates InRelease
    Temporary failure resolving 'de.archive.ubuntu.com'
  Reading package lists... Done
  W: Failed to fetch http://de.archive.ubuntu.com/ubuntu/dists/xenial/InRelease 
 Temporary failure resolving 'de.archive.ubuntu.com'
  W: Failed to fetch 
http://de.archive.ubuntu.com/ubuntu/dists/xenial-updates/InRelease  Temporary 
failure resolving 'de.archive.ubuntu.com'
  W: Failed to fetch 
http://security.ubuntu.com/ubuntu/dists/xenial-security/InRelease  Temporary 
failure resolving 'security.ubuntu.com'
  W: Some index files failed to download. They have been ignored, or old ones 
used instead.

  It should be corrected to return useful exit code, so that scripts can
  take the appropriate actions ...

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1693900/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1907850] Re: Cache not generated for all translations

2021-03-05 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Groovy)
   Status: Triaged => In Progress

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1907850

Title:
  Cache not generated for all translations

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Confirmed
Status in apt source package in Bionic:
  Confirmed
Status in apt source package in Focal:
  Confirmed
Status in apt source package in Groovy:
  In Progress
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  In bug 1161743 we discovered that if a system is configured with multiple 
locales, only the locales of the user who generated the apt-cache will be 
available for translated descriptions.

  [Test case]
  An automated test case has been included as a snippet in  
test/integration/test-bug-49-search-in-all-translations which runs during 
autopkgtest. That's sufficient for validation.

  Manual test:

  # apt install locales-all # get the locale
  # export LANG=sv_SE.UTF-8
  # locale
  LANG=sv_SE.UTF-8
  LANGUAGE=
  LC_CTYPE="sv_SE.UTF-8"
  LC_NUMERIC="sv_SE.UTF-8"
  LC_TIME="sv_SE.UTF-8"
  LC_COLLATE="sv_SE.UTF-8"
  LC_MONETARY="sv_SE.UTF-8"
  LC_MESSAGES="sv_SE.UTF-8"
  LC_PAPER="sv_SE.UTF-8"
  LC_NAME="sv_SE.UTF-8"
  LC_ADDRESS="sv_SE.UTF-8"
  LC_TELEPHONE="sv_SE.UTF-8"
  LC_MEASUREMENT="sv_SE.UTF-8"
  LC_IDENTIFICATION="sv_SE.UTF-8"
  LC_ALL=
  # apt update
  # apt-cache show tasksel | grep Desc
  Description-sv: tool for selecting tasks for installation on Debian systems
  Description-md5: cbbb747708986d11ea77c80b9b038fec
  # apt-cache showpkg tasksel
  Package: tasksel
  Versions:
  3.34ubuntu16 
(/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages)
   Description Language:
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages
    MD5: cbbb747708986d11ea77c80b9b038fec
   Description Language: sv
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_i18n_Translation-sv
    MD5: cbbb747708986d11ea77c80b9b038fec
   Description Language: en
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_i18n_Translation-en
    MD5: cbbb747708986d11ea77c80b9b038fec
  [...]

  So far so good, but now assume the root user actually has C configured
  as locale, and e.g. runs apt-cache show (or apt-daily.service does an
  update):

  root@g:~# rm /var/cache/apt/*.bin
  root@g:~# LANG=C apt-cache  show tasksel
  [...]
  Description-en: tool for selecting tasks for installation on Debian systems
   This package provides 'tasksel', a simple interface for users who
   want to configure their system to perform a specific task.

  root@g:~# apt-cache  showpkg tasksel
  Package: tasksel
  Versions:
  3.34ubuntu16 
(/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages)
   Description Language:
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_binary-amd64_Packages
    MD5: cbbb747708986d11ea77c80b9b038fec
   Description Language: en
   File: 
/var/lib/apt/lists/archive.ubuntu.com_ubuntu_dists_groovy_main_i18n_Translation-en
    MD5: cbbb747708986d11ea77c80b9b038fec

  This should show the sv locale as well given that it's still around
  (also we are still running with LANG=sv_SE.UTF-8), but it only
  generated the cache with the english language description in here.

  [Where problems could occur]
  People might end up seeing a wrong translation if other code paths are wrong, 
but then they'd already see them if they explicitly configured multiple 
languages and then switched the preferred one.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1907850/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1916050] Re: Invalid base64 for high-bit characters

2021-03-05 Thread Julian Andres Klode
** Changed in: apt (Ubuntu Groovy)
   Status: Triaged => In Progress

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apt in Ubuntu.
https://bugs.launchpad.net/bugs/1916050

Title:
  Invalid base64 for high-bit characters

Status in apt package in Ubuntu:
  Fix Released
Status in apt source package in Xenial:
  Confirmed
Status in apt source package in Bionic:
  Confirmed
Status in apt source package in Focal:
  Confirmed
Status in apt source package in Groovy:
  In Progress
Status in apt source package in Hirsute:
  Fix Released

Bug description:
  [Impact]
  The Base64Encode function incorrectly encodes using char instead of uint8_t, 
leading to characters with the high bit set to be encoded wrongly. This 
function is used for base authentication, and high bits are set for UTF-8 
character sequences, so it's likely affecting people with utf-8 user names and 
passwords and might cause them to be denied access to their https resources 
(though arguably they'd work around that).

  [Test case]
  A unit test has been added that checks that the encoding is correct.

  [Where problems could occur]
  This only affects the base64 encoding function, which is only used for 
quoting user names and passwords in basic auth in http, but it's also exposed 
to library users and like in python-apt, so problems could occur wherever they 
use that base64 encoding function as well.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1916050/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


  1   2   3   4   5   6   7   8   9   10   >