Bug#1064270: sysstat: FTBFS with old systemd (.sleep file doesn't get installed)

Package: sysstat Version: 12.7.5 Tags: patch Since systemd moved to /usr/lib, the old /lib/systemd install rule didn't match anymore. However, the fix in https://bugs.debian.org/1057421 was actually misguided. It fixed the "not found error", but it was still wrong. It caused the package to not com

Bug#1056639: FTBFS: requires network during mandatory build steps

Package: rust-urlshortener Version: 3.0.2-1 Severity: grave Justification for severity: Policy violation During a rebuild of Debian testing, we ran into this test failing during an sbuild controlled build on a VM with limit network connectivity (only configured APT repositories usable). See log e

Bug#1050236: FTBFS: 30 - testlpe fails

Hi. Two notes: 1) This doesn't just fail on rscv64 but also on amd64 and i386 with the same error as in the linked build log from the original report: [...] [ RUN ] LPETest.MeasureSegments_multi_px_1_0_2 add_actions_edit_document: no app! add_actions_pages: no app! add_actions_undo: no app!

Bug#1055729: FTBFS: faad2 fails to build on i386, blocking DSA fix transition

Severity: serious Package: faad2 Tags: patch upstream security fixed-upstream Please apply upstream fix https://github.com/knik0/faad2/commit/1001f9576cbb29242671c489cd861de61cfe08e2?diff=unified Kind regards, Sven

Bug#1055023: FTBFS: libpoe-component-client-dns-perl requires DNS during tests.

Package: libpoe-component-client-dns-perl Severity: serious Version: 1:1.054-3 While rebuilding debian trixie in a network-restricted environment, we ran into this build error (using sbuild): Can't call method "rdstring" on an undefined value at t/06_hosts.t line 143. # Looks like your test exite

Bug#1036174: FTBFS with nocheck option

Package: python-biopython Version: 1.80+dfsg-4 When doing a build via pbuilder/sbuild and setting the `nocheck` profile, the build-dependency on hevea is disabled (!nocheck). However, it is still executed. Log excerpt below: cd Doc && make make[2]: Entering directory '/<>/Doc' sed "s/\@VERSION\@/

Bug#1017811: node-wikibase-cli: FTBFS without Internet access

Interestingly, in our build environment which has some extremely limited network access, the package build fails before tests are even run: Linking /usr/bin/wb-set-label to /usr/share/nodejs/wikibase-cli/bin/wb-set-label Linking /usr/bin/wb-sparql to /usr/share/nodejs/wikibase-cli/bin/wb-sparql Li

Bug#778950: Fixed in shadow 1:4.12.3+dfsg1-3

Fixed 778950 1:4.12.3+dfsg1-3 thanks 1:4.12.3+dfsg1-3 includes the upstream patch for this (see upstream changelog inside the package) Cheers, Sven

Bug#1004598: fix available upstream

Audacity 3.2.0 is available upstream, fixing this.

Bug#1020580: FTBFS: ruby3.1 needs network/DNS during tests

Package: ruby3.1 Version: 3.1.2-3 The package violates policy 4.9, namely: For packages in the main archive, required targets must not attempt network access, except, via the loopback interface, to services on the build host that have been started by the build. The relevant snippet from a build

Bug#1017862: scipy: FTBFS on 32bit arches fixed

fixed 1017862 1.8.1-8 thanks Upload of 1.8.1-8 contained https://salsa.debian.org/python-team/packages/scipy/-/blob/master/debian/patches/pythran_type_PR16646.patch - which fixes this both according to the original failure report and my build of scipy for i386. Cheers, Sven

Bug#1016678: dose3: Overly sensitive to newlines in package control fields

Package: dose3 Version: 7.0.0-1 Tags: patch dpkg, apt, grep-dctrl and other tools handle multi-line Depends, Suggests, etc. fields just fine, even though policy only defines Description as a multi-line field (and dpkg-gencontrol folds such multiline fields in a source's debian/control into single

Bug#1016357: Fails tests with insufficient network connectivity

Package: golang-github-pion-turn.v2 Version: 2.0.8-2 Severity: Serious The package fails tests (see log of failing test below) when run on a machine with limited network (in this case: Can only reach the relevant APT repository server). This is a policy violation, which says that package tests mus

Bug#1016356: FTBFS: unpaper fails to build against ffmpeg 5.0

package: unpaper Version: 6.1-2 Severity: serious Build log excerpt below. There is a new upstream version, but I'm not sure if that builds cleanly against the new ffmpeg. ffmpeg version: 7:5.0.1-3 (lbavformat-dev and friends) dpkg-buildpackage - Command: dpkg-buildpackage --sa

Bug#1015710: FTBFS: click fails when builder has restricted network (misses mock)

Package: click Version: 0.5.0-7 Short story: Package needs python3-mock due to a wrong comparison in upstream setup.py, can't install it without network and then fails to build in the test phase. At work, I'm rebuilding all packages in Debian testing. click fails there, during the test phase. And

Bug#1012685: reported bug is in error. tilix is installable.

The bug report is apparently erroneously reported against tilix version 1.9.5-1 - which actually fixed the issue, as far as I can tell. Version 1.9.5-1 is installable (no "impossible dependency") in debian/sid

Bug#1011875: golang-github-opencontainers-runtime-tools: FTBFS unreproducible

Hi there. Just a side note: The test fails on my test builds (with the same error as in the original report. Because it relies on having a working network connection. Which it shouldn't do. (my test builds are on a connection that only allows access to a Debian mirror). Cheers, Sven

Bug#1013304: FTBFS if BUILD_DIR is set.

Package: scrcpy Version: 1.23-1 During rebuilds of debian-testing, scrcpy failed to build. Debugging this, it turns out that this is due to the fact that the environment already had BUILD_DIR set (to "build"). However, debian/rules expects that it is unset (indirectly): It expects that server/bui

Bug#1009643: Acknowledgement (Puppet: Fails to work with Ruby 3.0)

Just a quick update: The problem of this particular one is: in puppet/file_system.rb def self.symlink(path, dest, options = {}) @impl.symlink(assert_path(path), dest, options) end Changing this to: def self.symlink(path, dest, **options) @impl.symlink(assert_path(path), dest, **o

Bug#1009643: Puppet: Fails to work with Ruby 3.0

Package: puppet Version: 5.5.22-4 Severity: serious Justification: Fails at core functions Since testing switched /usr/bin/ruby to use ruby3.0, a previously working Puppet manifest fails: s...@larsa.muc:/tmp$ ruby3.0 /usr/bin/puppet apply t.pp Notice: Compiled catalog for larsa.muc.corp.google.co

Bug#1008517: FTBFS: ruby-html-proofer requires network during tests

Package: ruby-html-proofer Version: 3.19.2-4 During rebuilds of debian-testing on our build systems (which have minimal network available, in particular, they can only reach the local APT mirror, we ran into the failure below. IMHO, generally disabling checks via 'nocheck` in such an environment i

Bug#1005745: fai-setup-storage obfuscates parted error.

Package: fai-setup-storage Version: 5.10.3 At https://github.com/faiproject/fai/blob/dc83bb502a9af5666fa4344ba8b710bbbd5c827d/lib/setup-storage/Exec.pm#L97 fai-setup-storage turns an "overlapping partitions" error into an "internal error". The problem we ran into with this is that this also appe

Bug#1003969: sudo: FQDN option not overridable early on.

9, which means the patch I mentioned before has been integrated. Cheers, Sven Am Sa., 22. Jan. 2022 um 16:13 Uhr schrieb Marc Haber : > > tags 1003969 confirmed pending - patch > thanks > > On Wed, Jan 19, 2022 at 01:29:19PM +0100, Sven Mueller wrote: > > Tags 1003969 + f

Bug#1003969: sudo: FQDN option not overridable early on.

warning is shown when sudo is used, but it remain usable. Cheers, Sven Am Mi., 19. Jan. 2022 um 13:29 Uhr schrieb Sven Mueller : > > Tags 1003969 + fixed-upstream patch confirmed > Thanks > > https://www.sudo.ws/repos/sudo/rev/8c6eaa503793 has the upstream patch.

Bug#1003969: sudo: FQDN option not overridable early on.

Tags 1003969 + fixed-upstream patch confirmed Thanks https://www.sudo.ws/repos/sudo/rev/8c6eaa503793 has the upstream patch.

Bug#1003969: sudo: 1.9.8p2-1 audit plugin breaks if hostname doesn't resolve

Package: sudo Version: 1.9.8p2-1 Severity: important This is a regression compared to 1.9.5. Reproduction: As root, run: sudo -u nobody bash Expect: # sudo -u nobody nobody@foobarxyz:/root$ Does: # sudo -u nobody bash sudo: unable to resolve host foobarxyz: Name or service not known sudo: er

Bug#1003870: FTBFS: ilorest - missing dep on python3-pbr

Package: ilorest Version: 3.2.2+ds1-2 Severity: serious When rebuilding ilorest, I ran into error below on my build server which has limited network access. AFAICT, adding a build-dependency on python3-pbr should address this failure. Cheers, Sven + python3.9 setup.py install --install-layout=de

Bug#1001785: texlive-extra affected by log4j CVEs

rieb am Sa., 18. Dez. 2021, 14:47: > Am 16.12.2021 um 09:38 teilte Sven Mueller mit: > > Hi Sven, hi Norbert, > > > texlive-extra-utils contains arara (https://github.com/islandoftex/arara > ) > > which was updated two days ago via TeX Live ( > https://www.tug.org/

Bug#1001785: texlive-extra affected by log4j CVEs

Package: texlive-extra-utils Severity: grave Version: 2021.20211127-1 Tags: security texlive-extra-utils contains arara (https://github.com/islandoftex/arara) which was updated two days ago via TeX Live (https://www.tug.org/texlive/) which was updated slightly after that. Please update to the newe

Bug#1001209: dnsmasq: Regression: Fails to direct queries to 127.0.1.1#10053

a breaking change for existing installations, I don't think that would be appropriate. Cheers, Sven Michael Biebl schrieb am Do., 9. Dez. 2021, 19:45: > On Mon, 6 Dec 2021 14:20:51 +0100 Sven Mueller > wrote: > > Package: dnsmasq > > Version: 2.86-1.1 > > Severity

Bug#1001209: quick update

This bug is actually caused by what appears to be an accidental change: --local apparently is now an alias to --address while the manpage says it is an alias to --server. Still a regression and/or breaking change. Cheers, Sven

Bug#1001209: dnsmasq: Regression: Fails to direct queries to 127.0.1.1#10053

Package: dnsmasq Version: 2.86-1.1 Severity: Serious We have this our /etc/NetworkManager/dnsmasq.d/local-proxy: local=/in.ourdoma.in/127.0.1.1#10053 This leads to: dnsmasq: Bad address in --address at line 2 of /etc/NetworkManager/dnsmasq.d/goobuntu-dnsproxy The syntax used is exactly as docu

Bug#1001003: pbseqlib source declares Architecture:any but is only buildable for 64bit archs

Source: pbseqlib Version: 5.3.5+dfsg-2 I do rebuilds of packages in testing. I schedule them for amd64 and i386 depending on what the DSC reports in "Architecture:". For pbseqlib, this header reports: "any" However, pbseqlib depends on libpbcopper-dev which is only available for 64 bit architectu

Bug#996387: FTBFS - also with ruby 2.7.4

Hi. I get the same "rpc/rpc.h" error while rebuilding against (at least essentially) debian-testing and ruby 2.7.4. gcc -fdebug-prefix-map=/<>=. -I. -I/usr/include/x86_64-linux-gnu/ruby-2.7.0 -I/usr/include/ruby-2.7.0/ruby/backward -I/usr/include/ruby-2.7.0 -I. -Wdate-time -D_FORTIFY_SOURCE=2

Bug#994204: Upgrading severity

severity 994204 important found 994204 13.5.2 thanks This breaks many use cases, as already described by Arnaud and Fabian. It restarts services which explicitly asked _not_ to be restarted, please fix. Or at least revert https://salsa.debian.org/debian/debhelper/-/commit/6067bc2 until the problem

Bug#987234: hard dependency on python version can conflict with python-gi deps

Package: gstreamer1.0-python3-plugin-loader Version: 1.18.3-1 When building gst-python1.0, it creates a dependency on libpython3.x for the binary package gstreamer1.0-python3-plugin-loader. However, that binary also depends on python3-gi. We are rebuilding all packages we ingest from Debian. It so

Bug#984816:

ed by Sven Mueller) Signed-off-by: Sven Mueller --- busybox_1.30.1/klibc-utils/resume.c +++ busybox_1.30.1-new/klibc-utils/resume.c @@ -105,6 +105,13 @@ } ofs = (argv[1] ? xstrtoull(argv[1], 0) : 0); + if (ofs != 0ULL && + (fd = xopen("/sys/power/resume_offset", O_WRONLY

Bug#984816:

Tags 984816 + patch upstream Severity 984816 important This is https://bugs.busybox.net/show_bug.cgi?id=12006 - which also has a patch which wasn't adopted yet (but is straight from klibc-utils, so really should get adopted). Setting severity to important, as this makes hibernation to a swapfile

Bug#984816: busybox resume fails to resume with swap file after hibernation

Package: busybox-static Version: 1:1.30.1-6 Hi. I wasn't able to figure out all the details yet and likely won't get to that in the next few weeks. However, I tried getting hibernation to work on a machine with only a swap file. This failed miserably (machine appeared to hibernate properly, but o

Bug#984769: FTBFS: python-escript fails if DFLAGS is set

I should add: dpkg-buildflags manpage does not list DFLAGS as a supported flag, but explicitly mentioned that more flags might get added in the future if needed for more languages. DFLAGS was added with commit d01d1f49c9734a39fc278f4d5b75444c96491023 Author: Guillem Jover Date: Fri Nov 27 01:

Bug#984769: FTBFS: python-escript fails if DFLAGS is set

Package: python-escript Version: 5.6-2 In my build enviroments, the DFLAGS dpkg-buildflags option is set (to '-frelease'). python-escript in site_cons/extractdebbuild.py does not know about this flag but should actually ignore it, AFAICT. Suggested fix: Add DFLAGS to the ignoreflags list in line

Bug#983470: cdbs: waf-unpack needs Python 2.7

Am Mi., 24. Feb. 2021 um 19:34 Uhr schrieb Jonas Smedegaard : > severity -1 normal > > Hi Sven, > > Quoting Sven Mueller (2021-02-24 19:02:41) > > /usr/lib/cdbs/waf-unpack is using /usr/bin/python and it's syntax is > > Python 2 only. > > I have no clue

Bug#983472: dee: library examples use Python 2.7

Package: libdee-1.0-4 Version: 1.2.7+17.10.20170616-6 libdee-1.0-4:amd64: /usr/share/doc/libdee-1.0-4/examples/master-model.py libdee-1.0-4:amd64: /usr/share/doc/libdee-1.0-4/examples/slave-model.py Both use /usr/bin/python as their interpreter and use Python 2 syntax. Python 2 is EOL and https:/

Bug#983470: cdbs: waf-unpack needs Python 2.7

Package: cdbs Version: 0.4.163 Severity: Serious /usr/lib/cdbs/waf-unpack is using /usr/bin/python and it's syntax is Python 2 only. I have no clue what a WAF is, so I didn't spend time trying to fix this issue, sorry. Marking as serious as Python 2.7 is EOL and https://wiki.debian.org/Python/2Re

Bug#983468: python3-newt examples use /usr/bin/python

Package: python3-newt Version: 0.52.21-4 https://wiki.debian.org/Python/2Removal Says packages should not use /usr/bin/python While these examples are not an essential part of the package, they probably should get fixed. python3-newt:amd64: /usr/share/doc/python3-newt/examples/popcorn.py python3-

Bug#983092: xapps-common:all is not properly arch-independent

ioned library package I spoke about. Doh. Cheers, Sven Andreas Henriksson schrieb am Fr., 19. Feb. 2021, 21:22: > Hello, > > On Fri, Feb 19, 2021 at 10:45:41AM +0100, Sven Mueller wrote: > > Package: xapp > > Version: 2.0.6-1 > > Severity: serious > > > > xap

Bug#983092: xapps-common:all is not properly arch-independent

Package: xapp Version: 2.0.6-1 Severity: serious xapps-common is tagged as architecture:all, but the generated xapp-sn-watcher.desktop included in it depends on the architecture it was built on. By accident, we rebuilt the arch:all packages from xapp on i386, while our main architecture is amd64.

Bug#982677: FTBFS - required network connection

Hi, thanks for the background and for cross-checking. As it seems, this was indeed a problem with a too old puppetlabs-ring-middleware-clojure package being pulled in during the build. Not sure why (and why I missed that when comparing the used build dependencies). Kind regards, Sven Am So., 14.

Bug#982677: FTBFS - required network connection

Severity: serious Package: rbac-client-clojure Version: 0.9.4-2 TTBOMK, Policy says that builds shouldn't rely on networking. However, when attempting to build rbac-client-clojure on my (well, mostly) network-less builder VM, I run into this error: debian/rules override_dh_auto_build make[1]:

Bug#981870: FTBFS: Doesn't build against ruby-nokogiri 1.11.1

Package: ruby-azure-storage-common Version: 2.0.1-4 Severity: Serious This makes the package unbuildable in both unstable and testing (bullseye). Error message during build: /usr/lib/ruby/vendor_ruby/rubygems/dependency.rb:309

Bug#980787: libtss2-esys split causes regression

Severity: Grave Package: tpm2-tss Version: 3.0.3-1 Machines upgrading from previous versions lose access to the TPM because libtss2-tcti-device0 (at least, not sure when others would be needed) are not automatically installed during upgrade. Neither does tpm2-tools depend on these dynamically load

Bug#979252: New gpsd migrated to testing

Hi. Just wanted to let you know that the new gpsd is now in testing and that foxtrotgps now FTBFS in testing (and probably unstable), as best I can tell, for the very reason mentioned here. There are also lots of warnings about deprecated types (IIUC), but those don't make the build fail )(yet?).

Bug#972912: [Pkg-samba-maint] Bug#972912: Incorrect fix to bug

schrieb Mathieu Parent < math.par...@gmail.com>: > Le lun. 21 déc. 2020 à 16:06, Sven Mueller a > écrit : > > > > Well, turns out that apparently nobody else bothers with a .symbols file > for Python extensions. I looked at the packages for samba, numpy, mypy and >

Bug#979327: xalan: FTBFS on i386: architecture confusion

No worries. Thanks for the quick response. Am Di., 5. Jan. 2021 um 18:05 Uhr schrieb Bill Blough : > I had already pushed the initial fix by the time I received this, so I've > reopened the bug. I'll address it later this evening. >

Bug#979327: xalan: FTBFS on i386: architecture confusion

That might work, but I think if you really want to go without overriding the build directory (which is easiest), you should actually use DEB_BUILD_GNU_TYPE - the target gnu type, not the host gnu type (which are the same normally, but different when cross-building). See https://wiki.debian.org/Cros

Bug#979327: xalan: FTBFS on i386: architecture confusion

Package: xalan Version: 1.12-4 This was previously reported and should have been fixed in 1.12-3, according to the changelog, however, I see the error below when trying to build for i386 (see also https://bugs.debian.org/976299#10). The problem appears to be that when building for i386, the DEB_TA

Bug#979243: [Pkg-rust-maintainers] Bug#979243: FTBFS: rust-cpp-demangle - obsolete dep on rust-if-cfg-0.1-dev

Hi Peter. Odd. We are rebuilding packages in debian testing, but for some unknown reason (I'm still looking into it), the rust-cfg-if-0.1 source package was not considered. After manually adding it, everything works fine, thanks for the hint. Cheers, Sven Am Mo., 4. Jan. 2021 um 18:34 Uhr schrie

Bug#979243: FTBFS: rust-cpp-demangle - obsolete dep on rust-if-cfg-0.1-dev

Package: rust-cpp-demangle Version: 0.2.13-1 Severity: Serious It has this build dependency: librust-cfg-if-0.1+default-dev (>= 0.1.9-~~) , (And a matching, hard-coded Depends line on the binary package) However, this version of rust-cfg-if is now gone and replaced by 1.0 (Just FYI: the same h

Bug#972912: [Pkg-samba-maint] Bug#972912: Incorrect fix to bug

ebian/changelog? Cheers, Sven Am Do., 17. Dez. 2020 um 16:55 Uhr schrieb Mathieu Parent < math.par...@gmail.com>: > Le jeu. 17 déc. 2020 à 16:48, Sven Mueller a > écrit : > > > > Hi Mathieu. > > Hi, > > > Just wanted to say that your fix here seems wrong:

Bug#971216: doxygen build error

riginal dh_doxygen script? CC'ing Paolo for this, see intro. Cheers, Sven Attached is a patch to the included dh_doxygen Description: Remove custom dh_doxygen (broken with newer doxygen) Once upon a time, the custom script was useful, but this is not the time. Author: Sven Mueller https://bugs

Bug#971216: doxygen build error

Hi there. I believe this error is solely caused by embedding dh_doxygen into the imagemagick source package instead of using the dh_doxygen helper from the doxygen package. I tried a build with the debian/scripts/dh_doxygen call in debian/rules replaced by /usr/bin/dh_doxygen and it succeeded. C

Bug#972912: Incorrect fix to bug

Hi Mathieu. Just wanted to say that your fix here seems wrong: The symbols file says when a specific symbol for a specific lib was added. If I rebuild ldb against Python 3.9, it will suddenly claim that - for example - symbol PYLDB_UTIL_2.1.0@PYLDB_UTIL_2.1.0 was added to the package - for the Py

Bug#977542: golang-github-revel-revel - Depends on network in tests

Package: golang-github-revel-revel Version: 1.0.0-1 golang-github-revel-revel fails to run tests if the network doesn't provide working DNS. Technically, this is a policy violation. Please disable the relevant test or make it independent of the network. === RUN TestGetCustom --- FAIL: TestGetCu

Bug#977038: Confirmed: build-dep on libgrpc-java fixes the issue

declared dependency on libgrpc-java. 0.24.0-1 does have that dependency. Sorry for the noise. Am Fr., 11. Dez. 2020 um 10:41 Uhr schrieb Sven Mueller < sven.muelle...@gmail.com>: > Hi Tony. > > You might be right and I'll try to debug a little more. > I'll share the bui

Bug#977038: Confirmed: build-dep on libgrpc-java fixes the issue

0. Dez. 2020 um 18:45 Uhr schrieb tony mancill : > On Thu, Dec 10, 2020 at 04:51:53PM +0100, Sven Mueller wrote: > > Hi. > > > > I just re-ran tests after adding a build dependency on libgrpc-java and > > that fixed the reported issue. > > Hi Sven, > > Woul

Bug#977038: Confirmed: build-dep on libgrpc-java fixes the issue

Hi. I just re-ran tests after adding a build dependency on libgrpc-java and that fixed the reported issue. Cheers, Sven

Bug#977038: google-oauth-client-java: FTBFS: Failed to execute goal on project google-oauth-client: Could not resolve dependencies for project com.google.api-client:google-api-client:jar:1.27.1: Canno

Package: google-oauth-client-java Version: 1.28.0-1 Severity: serious This can likely be fixed with a build dependency on grpc-java [ERROR] Failed to execute goal on project google-oauth-client: Could not resolve dependencies for project com.google.oauth-client:google-oauth-client:jar:1.28.0: Can

Bug#974065: [DRE-maint] Bug#974065: rails: nocheck builds should not depend on qunit-selenium

Uhr schrieb Antonio Terceiro < terce...@debian.org>: > On Mon, Nov 09, 2020 at 03:40:15PM +0100, Sven Mueller wrote: > > Package: rails > > Version: 2:6.0.3.4+dfsg-1 > > Tags: patch > > > > Turns out that this isn't needed and pulls in a lot of dependencie

Bug#972552: Plan forward?

Hi. Just wanted to check in and ask what the plan going forward is. IIUC: The apt version in Stable already has the change that turned the warning into an error (the change essentially rolled back with https://salsa.debian.org/apt-team/apt/-/commit/998a17d7e6f834c341f198ca5b6df2f27e18df38 ). Th

Bug#975045: [Debian-zh-dev] Bug#975045: galternatives fails to build for binnmu

well. (We also would tag the version numbers similarly if we have a local patch, but IIRC, that wasn't the case here) Yangfl schrieb am Mi., 18. Nov. 2020, 16:42: > Sven Mueller 于2020年11月18日周三 下午6:39写道： > > > > Package: galternatives > > Version: 1.0.7 > > Tags: pat

Bug#975045: galternatives fails to build for binnmu

Package: galternatives Version: 1.0.7 Tags: patch debian/rules compares the version from debian/changelog with the version reported by the INFO dict. However, this fails if a binnmu is at play: A binnmu adds a changelog entry with +bX (X being an integer), but generally speaking, rebuilds might a

Bug#975041: FTBFS: galternatives fails during "clean"

Quick update: I wanted to open a pull request on Salsa, but accidentally actually submitted the fix directly. So whenever you do a new upload, it will include that fix. commit db478772da895ffb2b9bf91adb12903c3d440f9c https://salsa.debian.org/chinese-team/galternatives/-/commit/db478772da895ffb2b9b

Bug#975041: FTBFS: galternatives fails during "clean"

ixes this. Description: Make galternatives depend on gir1.2-gdk3.0, needed to run "clean" Author: Sven Mueller Last-Update: 2020-11-18 --- This patch header follows DEP-3: http://dep.debian.net/deps/dep3/ Index: galternatives-1.0.5/debian/control ===

Bug#974065: rails: nocheck builds should not depend on qunit-selenium

Package: rails Version: 2:6.0.3.4+dfsg-1 Tags: patch Turns out that this isn't needed and pulls in a lot of dependencies if the build is done without tests. The attached patch moves the start of the redis server (only needed for tests) inside the override_dh_auto_test section in debian/rules (I s

Bug#969458: perl crash in eval command trying a failing require statement

Hi there. Any chance this could get fixed in the near future? Cheers, Sven

Bug#971700: debootstrap: Please add option to get full log on stderr

allows the caller of debootstrap to request that stderr is passed to it instead of being swallowed into debootstrap.log Author: Sven Mueller Forwarded: no Last-Update: 2020-10-02 --- This patch header follows DEP-3: http://dep.debian.net/deps/dep3/ Index: debootstrap-1.0.123/debootstrap

Bug#970443: mira: FTBFS due to race condition during doc build

Ooops, sorry, forgot the attachment. Am Mi., 16. Sept. 2020 um 14:58 Uhr schrieb Andreas Tille : > Hi Sven, > > On Wed, Sep 16, 2020 at 02:49:54PM +0200, Sven Mueller wrote: > > Package: mira > > Version: 4.9.6-4 > > Tags: patch > > Thanks a lot for your invest

Bug#970443: mira: FTBFS due to race condition during doc build

Package: mira Version: 4.9.6-4 Tags: patch When building concurrently, the doc/docbook/*_part.html files can fail to generate. The reason is - as far as I can tell, that the %_part.html rule in doc/docbook/Makefile.in doesn't list versionfile as a dependency, but several of the _part.xml files act

Bug#970434: mira: FTBFS with current docbook

Looks like the error isn't caused by docbook. Trying to track this down some more and hopefully also provide a fix.

Bug#970434: mira: FTBFS with current docbook

Package: mira Version: 4.9.6-4 Making all in docbook make[3]: Entering directory '/<>/doc/docbook' Making versionfile Making .html out of _part.xml chap_bitsandpieces_part.html Making .html out of _part.xml chap_mirautils_part.html Making .html out of _part.xml chap_est_part.html Making .html out

Bug#970379: FTBFS: cpp-hocon fails to build against boost1.71

Package: cpp-hocon Version: 0.1.7-1 Excerpt from build log: [100%] Linking CXX executable ../../bin/libcpp-hocon_test cd /<>/obj-x86_64-linux-gnu/lib/tests && /usr/bin/cmake -E cmake_link_script CMakeFiles/libcpp-hocon_test.dir/link.txt --verbose=1 /usr/bin/c++ -Wextra -std=c++11 -Wall -Wno-unus

Bug#970317: plymouth can cause initrd creation to fail.

Package: plymouth Version: 0.9.4-3 We encountered a way how plymouth can cause initrd creation to fail: /usr/share/initramfs-tools/hooks/plymouth contains a call to plymouth-set-default-theme, reading the set theme name. plymouth-set-default-theme fails back to "text" apparently, if the desired t

Bug#969076: dpkg: Fails if package containing symlink to network FS is upgraded

Guillem Jover schrieb am Fr., 4. Sept. 2020, 00:28: > Hi! > > On Thu, 2020-08-27 at 09:19:52 +0200, Sven Mueller wrote: > > Package: dpkg > > Version: 1.19.7 > > Tags: patch > > > When a package contains a symlink that reaches into a filesystem that >

Bug#969076: dpkg: Fails if package containing symlink to network FS is upgraded

ck if it points to a directory. Treate ENOKEY like ENOENT, as the point of the check is to ensure that the new file doesn't end up in the directory the symlink could be pointing to. Author: Sven Mueller Last-Update: 2020-08-26 --- Index: d

Bug#960420: Needs versioned dependency on slop

Hi there. The solution mentioned in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=960420#10 (splitting slop into two packages) would actually be the right way to go. If maim were to automatically generate a versioned dependency on slop based on the version installed during build, you would sti

Bug#966375: apt: Please provide an option to make "transient network errors" more detectable.

Package: apt Version: 2.1.7 Priority: wishlist Hi. While managing systems, I sometimes want to make sure that an "apt update" actually fetches new index files. However, if the download fails with a 5xx error (and, I think if there already is an index file for that repository), then apt just outpu

Bug#961798: python-pyaudio: FTBFS for doc package

Tags: patch The attached patch fixes the issue. The problem is that libjs-jquery doesn't contain files relevant here anymore and uses symlinks itself instead. Which dh_linktree doesn't seem support. Switching to node-jquery and the relevant path fixes the build error Index: python-pyaudio-0.2.11/

Bug#961798: python-pyaudio: FTBFS for doc package

Package: python-pyaudio Version: 0.2.11-1.1 Severity: serious python-pyaudio fails to build from source in both testing and unstable with the following error: dh_installinit: warning: Compatibility levels before 10 are deprecated (level 9 in use) dh_perl -O--buildsystem=pybuild dh_link -O--

Bug#958460: FTBFS with no network

Package: appstream-generator Version: 0.8.1-1 Severity: serious This is a policy violation, as the build process must not rely on network accessibility. Error message: [139/147] /usr/bin/xsltproc --nonet --stringparam man.output.quietly 1 --stringparam funcsynopsis.style ansi --stringparam man.th

Bug#952244: qtcurve: FTBFS: dh_install: error: missing files, aborting

Hi. Looks like this is due to a missing build-dependency on the libx11-xcb-dev package. Adding that dependency makes the package build for me. Cheers, Sven

Bug#951637: merge request

Tags 951636 + patch Tags 951637 + patch thanks See https://salsa.debian.org/haskell-team/DHG_packages/merge_requests/4

Bug#951637: FTBFS if mail address can't be detected by git

Package: haskell-githash Version: 0.1.3.1-3 Severity: serious I was attempting to build haskell-githash on a system that had no usable network and thus also no FQDN. I do interpret this as an attempt to access the network during build, though this is debatable, so feel free to downgrade the severi

Bug#951636: FTBFS with "nocheck" build profile

Package: haskell-githash Version: 0.1.3.1-3 If a build is attempted with the nocheck profile, the git package is not a build dependency, but debian/rules uses git in the "clean" target even with the nocheck profile. The execution in the "clean" target either needs to be conditional on not having "

Bug#946274: gnuradio: Fails to build if build system has Python 3.6 installed

Source: gnuradio Version: 3.8.0.0-6 Severity: normal If the build system has Python 3.6 (as /usr/bin/python3.6) installed while the libpython3-dev packages uses Python 3.7, the build fails. This can still be the case on old systems if the user has any dependencies on older Python. This does not ha

Bug#946065: gnuradio: uses parameters with dh_auto_test that make ninja bail out

Source: gnuradio Version: 3.8.0.0-6 Severity: normal >From the buildd log: if case "amd64" in *"i386"*) false;; *) true;; esac; then HOME=/<>/debian/test dh_auto_test -- ARGS+='--timeout 10 -E systematic' ; fi cd obj-x86_64-linux-gnu && LC_ALL=C.UTF-8 MESON_TESTTHREADS=4 ninja test "ARGS+=--timeo

Bug#792224: node-oauth still fails during tests with no network

For better readability, here is the text content from the control message I sent (but which is hidden by default): On my build host without network, I still get: ♢ SHA1 Hashing When using the SHA1 Hashing function ✓ we get the specified digest as described in http://oauth.net/core/1.0/#s

Bug#942054: node-oauth does not honor "nocheck" build profile

Source: node-oauth Version: 0.9.11-1 Severity: normal Debian policy defines the "nocheck" build profile to enable skipping any tests. I would need to skip tests because the tests access the network (see http://bugs.debian.org/830214) and my build machine doesn't allow DNS resolution. The contents

Bug#930734: glibc: Doesn't honor noudeb build profile

Package: glibc Version: 2.28-10 Tags: patch https://wiki.debian.org/BuildProfileSpec documents noudeb specifically to disable building .udeb packages, but the glibc sources only honor the (deprecated!) stage1 profile. The attached patch disables building udebs if stage1 isn't given but noudeb is (

Bug#929442: Acknowledgement (qtwebsockets-opensource-src: debian/rules doesn't honor nocheck)

Merge request against master branch created at https://salsa.debian.org/qt-kde-team/qt/qtwebsockets/merge_requests/6

Bug#929442: qtwebsockets-opensource-src: debian/rules doesn't honor nocheck

Source: qtwebsockets-opensource-src Version: 5.12.3-1 (Also relevant for versions in unstable and testing, for what it's worth) Builds of qtwebsockets-opensource-src fail some test(s) due to the rather special network setup on the system. This would be easy to remedy by just building with the "no

  1   2   3   4   5   >