May 2022 - openSUSE Factory - openSUSE Mailing Lists

openSUSE Release Engineering meeting 04.05.2022
by Lubos Kocman 06 Jun '22

06 Jun '22

All meeting minutes can be found here: https://etherpad.opensuse.org/p/ReleaseEngineering-meeting Meeting is hosted here https://meet.opensuse.org/ReleaseEngineeringMeeting ## Attendees guillaumeg, DimStar, rbrown, lkocman, mpiala, ddemaio, maurizio (m4u).maxlin, Sarah ## Internal SUSE IT Supported distribution Guests: Martin Piala, Vojtech Pavlik Idea: openSUSE Leap as supported system by SUSE IT! Vadim G. is looking for a help to kickstart the effort. We're thinking of starting a hackweek with a Leap 15.4 spin with some preinstalled tools for SUSE environment. https://hackweek.opensuse.org/21/projects/opensuse-build-supported-by-suse-… Our feature tracker for any issue references. https://code.opensuse.org/leap/features/issue/77 Next steps collect requirements for Minimal Viable Product I believe Vojtech could help us there. How does Rancher product portfolio fit in? Datto is using gnome-initial-setup for post-install tweaks. https://github.com/endlessm/gnome-initial-setup/blob/master/HACKING We could start there as well. However, they're already looking for some futher management tools. I believe that we do not need any 3rd party tools. Perhaps SUMA could help with post-installation management. Martin: currently supported systems by IT are Windows 10 and MacOS, we would like to change this Martin: we'd like to have pre-installed tools required for daily work. slack, teams. We have company portals for Windows10 and Mac and some of the apps can be already preinstalled for the image. Martin: we do not yet have experience for responsibility of building Linux apps and we need your help. Ideally it should be done by the end of the quarter. End of July It's not decided yet what should we use. Whether we need a script post- install. Richard: I do have a couple of questions: Which HW are we talking about? * How often will be the recommended hardware changing? Martin: We have some predefined laptops specific. 2 hardware specs (standard, power). New ones are Lenovo, we're also rotating the older Dells. Martin will discuss this with team. In the next HW rotation we should take the support of the chosen OS into consideration. * Would the best option be Leap based? And perhaps risk some of the missing hardware support? * Or should we consider something Tumbleweed based? Perhaps you will end up with something like MicroOS * Desktop. Teams, Slack would be then consumed from Flatpaks. lkocman: I recommend to consider two approaches: provide root password to the user or not. For less technical users it would be preferred if SUSE it would do most of the changes. Martin: For IT are also preferred these two options, one with admin access (here the user is responsible for the setup), and another with limited permissions, where IT is responsible for maintenance + the laptop must be enrolled (most probably in VMware Workspace One, currently under research). ## Leap Leap Micro 5.2 is RC! We've also progressed on release-notes-leap-micro (Thanks cboltz and Lukas) https://doc.opensuse.org/release-notes/x86_64/openSUSE/LeapMicro/5.2/ Rework of LeapMicro get-o-o layout community worked on overview for Leap Micro https://etherpad.opensuse.org/p/weeklymeeting20220505#L51 We should also consider single page layout MicroOS and backport overview changes as well https://github.com/openSUSE/get-o-o/pull/87 A lot of changes from SLE 15 SP4 post-RC. Removing snapshot at this point in time was not so good choice in the end. Some more changes will be taken in today, tomorrow. Mass rebuild is planned in next few days (Friday/Saturday). Maint snapshot for SLE 15 SP4 on Sunday. We need to be able to finish the RC build, that's not possible now. Leap Desktop/Workstation workshop is next week! 50+ people have signed up for workshop https://www.reddit.com/r/openSUSE/comments/udr5ly/desktop_workshop_for_next… - syncup with Scott and Yi Fan prior the acutal workshop Looking for drivers who could help us "kickstart the initial workshop". Long story We need some people who already know what to do (how lucid works, create some initial topics) to inspire others. Interesting last minute feature for laptop customizations. (Big thanks Dirk) I'll see what would be easily adoptable. SLED is interested as well. https://code.opensuse.org/leap/features/issue/75 ## openSUSE Tumbleweed openSUSE:Factory build fail stats: 108 failed, 10 unresolvable (last week: 123/11) https://tinyurl.com/ysy4nnnz * GNOME 42.1 shipped in snapshot 0501 (snapshot 0429 and 0430 were blocked due to dracut) * KDE Plasma 5.24.5 staged - ready for either 0504 or 0505 * Meson 0.62 is staged - another round of 'syntax validation getting stricter' (not too many issues seen though) * GCC12 as default compiler still staged, results are looking very good (Staging QA all passed, waiting for the final go) lkocman: I hope that Alp will adopt to the new default(s) as well. Mentioned the llvm situation in Leap 15.X GPG Key to sign packages/repos is planned to be updated to be a 4096/RSA key (implementation schedule pending) Tracker bug : https://bugzilla.suse.com/show_bug.cgi?id=1199184 OBS seems to have some hard time finding workers for ceph and kernel builds: very frequently, staging projects are blocked by waiting for the 4 same packages still being scheduled (ceph, kernel-default, kernel-vanilla, kernel-debug) (constraints limit to 18 (ceph) resp 24 (kernels) workers) ## Richard (Kubic/MicroOS) Started Discussion with the Kubic community about winding down the Kubic project and focusing more attention on MicroOS (and MicroOS Desktop) - https://lists.opensuse.org/archives/list/kubic@lists.opensuse.org/thread/23… For those currently on Kubic, Kubernetes atop MicroOS is best addressed with using k3s/RKE direct from Rancher. Whether we need tooling/documentation to help with potential migrations is an open question. lkocman: This topic was raised also on the community meeting. Isn't Kubic the only Kubernetes and CNCF certified distro that we have now? Richard: depends on your definition of 'we' - CNCF certification is specific to the Kubernetes that is part of Kubic. Rancher k3s supports MicroOS and is itself CNCF certified, so 'we' will still have that. lkocman: The question that popped up was how to get the k3s running on Leap Micro. I did share the whitepaper for SLE Micro 5.1. Richard: this should just work with 5.2+. Objecting to the Leap Micro get.opensuse.org page overstating it's case at the expense of other openSUSE Projects that address the same use cases - https://github.com/openSUSE/get-o-o/issues/88 MicroOS (GNOME) Desktop has a fresh community effort to get out of beta, possibly by writing a fresh notification tool for transactional- updates instead of relying on PackageKit. Join the Telegram/Matrix MicroOS Desktop channel to join the effort. Podman 4.0.x now working perfectly Working on various interesting MicroOS research tracks: - kiwi-less/container first VM image building Status: POC - nerdctl support in toolbox Status: Broken prototype - buildkit and its limit as a container builder - TW-to-MicroOS conversion - curl-minimal and making it default (like Fedora) Status: Investigating - Re-arranging Factory so products like MicroOS can build with different PrjConf than Tumbleweed - MicroOS Desktop using TIU instead of transactional-update (ie. remove all individual rpm package management) (research not sure if it's worth updating docs yet) Please ping rbrown(a)suse.com if you are interested/have opinions on any of the above, feedback/thoughts/existing knowledge is welcome. ## Max Leap Micro * Have a look repotred Leap Micro selfinstall has different behavior compare to SLEM on openqa, it should boot iso instead of raw image for the testing, bsc#1197877 * Removed beta flag Leap 15.4 * Look into python-certbot bug, the culprit is: python-pyopenssl in SLE15 SP4 has been updated to version 20, it's requiring python- cryptography > 3.2 however we had 2.8 in SLE, the minimal requirement can not filled, this should affected many packages rely on pyopenssl, bsc#1198851 (There is a SLE maintenance request for python-cert*, but it's a major version bump) Marcus does not recommend forking package in Leap as there are other SLE python modules that will depend on it. * Removed snapshot repo * Preparing 15.4 RC build Lubos will send an annoucement that RC will be delayed because of incoming changes from SLE 15 SP4. * Build fail stats in Leap 15.4: 20 build fails, 0 unresolvables lkocman: I'll open individual bugs for remaining build failures. Santiago mentioned that he'll have more free cycles for Leap/openQA this week. I'll reach out to him regarding focus on the migration scenarios. ## Guillaume - Arm Tumbleweed: * Rolling again * Upcoming GCC 12 tested for aarch64: https://build.opensuse.org/project/show/home:dirkmueller:branches:openSUSE:… Leap 15.4: * aarch64: * Issue with Numpy/openBLAS fixed: https://bugzilla.suse.com/show_bug.cgi?id=1198885 * armv7: * Enlightenment package still fails in rpmlint check (missing whitelist?): https://bugzilla.opensuse.org/show_bug.cgi?id=1194047 lkocman: will reach out to simotek and Wolfgang * scc product class looks wrong (Marina/Marcus) (for all non-intel arches). x86_64 has OPENSUSE-BETA while for the others the class is SLES-ARM64-BETA, SLES-PPC-BETA, SLES-Z-BETA. https://sd.suse.com/servicedesk/customer/portal/1/SD-82355 (Internal, Ticket with SCC team, team lead is aware of the request). Request is now resolved. (lkocman: Correction trello shows me that 2/4 items are still pending). SCC team confirmed that this is done. I did check non- intel arches and they seem to all use OPENSUSE-BETA. Leap Micro 5.2: * SR to add Default-SelfInstall image for aarch64: https://build.opensuse.org/request/show/974751 Lubos will open a JIRA for 5.3. I'll ask if it would be option for 5.2 as well. lkocman to check if we could remove the firefox tests from the 42.X/15.0 upgrade test suite. It doesn't seem to happen outside of that (same for TW). lkocman will try to look into this ## Sarah - s390x SUSE has shutdown 3 LPARs because of performance issues Issue has been also reported to the CTO for ZSystems of IBM Tumbleweed: * Is rolling openSUSE Leap Beta: * no issues at the moment * maven issues are existing -> discussions about upgrading the Java version or how to fix it lkocman: slightly related topic, should we perhaps re-consider recommended memory amount in Release Notes? I think we still recommend something like 2G minimal / 4G or 8G optimal. * KDE for s390x adopted ## Doug * TSP * Solution seems to be found * No update on status on when it can be implemented * Community Meeting * Discussion/workshop planned for community meetings on May 10 & 12 to discuss successor of Leap * https://lists.opensuse.org/archives/list/project@lists.opensuse.org/thread/… * oSC22 * Will cancel all talks on confrimed by May 6 * Schedule mostly complete * Ordering badge and signage * Setting up venueless platform * Slides need to be 16x9 * OSCAL oSC22 * Will start creating schedule * Reached out to people to confirm their talks * GSoC * 4 proposals ranked for May 12 deadline Installfest in CZ. CfP deadline was Monday midnight: MicroOS Leap and Leap Micro D-Installer talks were submitted ## Dirk Not available lkocman: Thanks for the framework laptop setup We're not currently listed as framework laptop compatible distro https://gist.github.com/cornfeedhobo/8f32430dc6c52b9a4776df91701b1d96 * switched obs-arm-11 worker to aio=io_uring away from aio=threads which appears to reduce I/O cost (bsc#1197746), which caused issues on x86_64. still chasing that down. * did a full rebuild of openSUSE:Factory:ARM as Guillaume was away and openqa was broken anyway :) did introduce more issues than it fixed it appears * Still to work rust 1.59 / ghc issues on armv6 and armv7: all have the same root cause, use of SWP/SWPB instructions which is imperfectly emulated by the aarch64 host kernel. works fine with armv6/v7 kernel. * looked into gcc11 build failures In backlog: https://code.opensuse.org/leap/features/issue/17 ## Wolfgang (Package Hub), Scott Bahling Not available ## Maintenance team (Marina or Marcus, Maurizio (m4u)) - Marcus: 15.4 is set up channel wise, updates will go in testing occasionaly, if you need more ping me, no openqa setup yet lkocman: could you please help me with cleanup of maintenance tasks in https://progress.opensuse.org/projects/opensuse-leap-15-4/issues/gantt lkocman: will contact Jan Stehlik. Santi (PO for the topic) mentioned he'll work on it. lkocman: messaged Jan - Marcus: 15.3 openQA is ok again. lkocman: (still pending) will open a release notes entry for a change Security Advisory identification TODO Marcus: could you give me example of old and new values? - Marina and Maurizio: SLE-PackageHub overlaps (https://github.com/openSUSE/openSUSE-release-process/issues/71) work in progress following the initial list of overlapping packages (https://progress.opensuse.org/issues/106610#note-3) More incidents are under QA. Work is going on and under control (and tracked internally as jsc#MSC-303). Already released: (AH PERFECT!) SUSE:Maintenance:22929:265929 (babl libbabl-0_1-0 typelib-1_0-Babl-0_1) SUSE:Maintenance:23000:266006 (MozillaThunderbird enigmail) SUSE:Maintenance:23009:266004 (freerdp freerdp-devel freerdp-proxy libfreerdp2 libgsm) -> libgsm1 was missing in PH making freerdp not installable SUSE:Maintenance:22928:266007 (argyllcms csync libcsync0 libcsync- plugin-sftp libcsync-plugin-smb) SUSE:Maintenance:23015:267052 (gfbgraph-devel libgfbgraph-0_2-0 typelib-1_0-GFBGraph-0_2 librest gcr) -> librest, gcr and libgoa are needed too for making gfbgraph installable SUSE:Maintenance:23334:268316 (gutenprint gutenprint-devel) SUSE:Maintenance:23486:268867 (hp-drive-guard, upower) --> needs libupower-glib3 SUSE:Maintenance:23487:268897 (liba52-0 liba52-devel) SUSE:Maintenance:23506:268899 (libdvdread4 libdvdread-devel) Meet product security at the openSUSE Conference. There are about 3 talks on the maintenance and security topics. Come and say Hello! ## Adrian - OBS Not present feedback loop on mirrored submissions doesn't workstation. Lubos wills end an email with an example to Marco and Adrian. (Sent last week) Marco has reached out to me regarding the mirroring. He has example(s) from SUSE:SLE-15:SP4:GA

3 4

[opensuse-factory] new package mg for Factory
by Tomasz Hołubowicz 01 Jun '22

01 Jun '22

Dear Factory! I would like to submit a small package which could be really handy in need for those that are used to readline bindings or Emacs. This package seems to be available on other linux distributions already so imagine my surprise, when I did not find it available in openSUSE repositories. Mg is a micro Emacs clone, especially helpful when you want to edit something as sudoer and prefer to use something close to Emacs (not vim). As author of the package states on in package's README: "The intention is to be a small, fast, and portable Emacs-like editor for users who cannot, or do not want to, run the real Emacs for one reason or another. Compatibility with GNU Emacs is key for Mg, separating it from other ErsatzEmacs clones, because there should never be any reason to learn more than one Emacs flavor." Package has releases approximately once a year, is actively maintained and it shouldn't take long till the release of next version 3.6. I have still a lot to learn when it comes to openSUSE packaging (this would be my first packaged and submitted project) and I would welcome any feedback or suggestions where I could improve. Kind wishes Tomasz Hołubowicz

3 4

New Tumbleweed snapshot 20220530 released!
by Dominique Leuenberger 31 May '22

31 May '22

Please note that this mail was generated by a script. The described changes are computed based on the x86_64 DVD. The full online repo contains too many changes to be listed here. Please check the known defects of this snapshot before upgrading: https://openqa.opensuse.org/tests/overview?distri=opensuse&groupid=1&versio… Please do not reply to this email to report issues, rather file a bug on bugzilla.opensuse.org. For more information on filing bugs please see https://en.opensuse.org/openSUSE:Submitting_bug_reports Packages changed: boost-base boost-extra dpdk f2fs-tools (1.14.0.g67 -> 1.15.0) google-noto-fonts (20220516 -> 20220524) libdlm libidn libidn2 llvm14 (14.0.3 -> 14.0.4) mlterm (3.9.1 -> 3.9.2) python-Pillow (9.1.0 -> 9.1.1) python-matplotlib (3.5.1 -> 3.5.2) python-numpy (1.21.4 -> 1.21.6) python-packaging rubygem-ruby-dbus (0.18.0.beta6 -> 0.18.0.beta7) sshfs (3.7.2 -> 3.7.3) update-alternatives (1.21.1 -> 1.21.8) vim (8.2.4877 -> 8.2.5038) virtualbox virtualbox-kmp yast2-packager (4.5.4 -> 4.5.5) === Details === ==== boost-base ==== Subpackages: boost-license1_79_0 libboost_date_time1_79_0 libboost_filesystem1_79_0 libboost_iostreams1_79_0 libboost_locale1_79_0 libboost_program_options1_79_0 libboost_thread1_79_0 - Fix failing conversion of cpp_dec_float to double, depending on locale (gh#boostorg/multiprecision#464, boo#1199968). Add boost-mp-locale-fix.patch ==== boost-extra ==== - Fix failing conversion of cpp_dec_float to double, depending on locale (gh#boostorg/multiprecision#464, boo#1199968). Add boost-mp-locale-fix.patch ==== dpdk ==== - add kni-fix-build-with-Linux-5.18.patch to fix build with kernel 5.18 ==== f2fs-tools ==== Version update (1.14.0.g67 -> 1.15.0) - update to 1.15.0: * fsck.f2fs: fix broken file_map output * f2fs-tools: show segment/section layout correctly * dump.f2fs: compress: fix dstlen of LZ4_compress_fast_extState() * big endian fixes * cross compile fixes * mkfs.f2fs: set required quota types only * fsck.f2fs: Add progression feedback * fsck.f2fs: do not assert if i_size is missing i_blocks in symlink * f2fs-tools: separate other bugs in fsck_verify * fsck|dump.f2fs: add -M to get file map - drop f2fs-tools-1.4.0-bigendian.patch: obsolete ==== google-noto-fonts ==== Version update (20220516 -> 20220524) Subpackages: google-noto-fonts-doc noto-sans-fonts - Switch back to hinted ttf as unhinted otf causes blurring (boo#1199938) - Add obsoletes and provides for: - noto-mono-fonts: Got merged into noto-sans-mono-fonts - noto-sans-syriac* variants: Got merged into noto-sans-syriac-fonts - noto-sans-tibetan-fonts: Got renamed to noto-serif-tibetan-fonts - Update to version 20220524 - Updated Noto Sans Myanmar and Noto Sans Tangsa Fonts ==== libdlm ==== Subpackages: libdlm3 - Adjust BuildRequires libpacemaker-devel -> pkgconfig(pacemaker) ==== libidn ==== - Update keyring from source URL ==== libidn2 ==== Subpackages: libidn2-0 libidn2-0-32bit libidn2-lang - Refresh libidn2.keyring ==== llvm14 ==== Version update (14.0.3 -> 14.0.4) Subpackages: clang-tools clang14 clang14-doc libLLVM14 libc++-devel libc++1 libc++abi-devel libc++abi1 libclang-cpp14 libclang13 - Update to version 14.0.4. * This release contains bug-fixes for the LLVM 14.0.0 release. This release is API and ABI compatible with 14.0.0. - Don't use gold for linking anymore: on s390x we use ld.bfd with LLVMgold.so, on ppc64 we disable ThinLTO for now. - Using ld.bfd on s390x exposed an issue with the existing llvm_build_tablegen_component_as_shared_library.patch: linking llvm-tblgen with libLLVM.so means we also have to link libraries used for that (like LLVMTableGenGlobalISel) with libLLVM.so. - Rewrite summary and description for llvm-gold to point out that it can also be used with ld.bfd, recommend with binutils. - Prefer RPM macros over shell scripting, so that we can better inspect the build script with substitutions in place. - More memory for stage 1 build jobs due to recent OOMs. - Add %_libclang_sonum RPM macro to llvm-devel, since that might now diverge from %_llvm_sonum. - Rebase llvm-do-not-install-static-libraries.patch. ==== mlterm ==== Version update (3.9.1 -> 3.9.2) Subpackages: mlterm-canna mlterm-ibus mlterm-scim mlterm-wnn - version update to 3.9.2 * Support Fcitx5. * Support GTK4 for mlconfig. (./configure --with-gtk=4.0) * Support CSI > Ps q (XTVERSION). (Response: DCS>|mlterm(3.9.2)ST) * Add "MOSH_UDP_PORT" environmental variable to specify UDP port or port-range to communicate with a mosh server. * Add INSERT_CLIPBOARD shortcut key. (https://github.com/arakiken/mlterm/issues/19) * Add --fk / format_other_keys option (equivalent to "formatOtherKeys" of xterm), and change the default format of the escape sequence with modifyOtherKeys from CSI Ps; Ps u to CSI 27; Ps; Ps ~. * Add --with-iconv, --disable-daemon, --disable-split and --disable-zmodem options to ./configure script. * Add vte 0.66 API symbols to libvte compatible library. * Show server list dialog at startup on Android. * It is deprecated to use "mlterm" as the value of termtype / -y option. * 'termcap' config file stops to accept multiple terminal types concated by '|' (e.g. xterm|xterm-256) as key. * Arabic ligatures (U+644-U+622, U+644-U+622, U+644-U+625, U+644-U+627) occupy two logical columns instead of one even if --dyncomb option isn't specified. * KBD_INPUT_NUM and MOUSE_INPUT_NUM environmental variables (for mlterm-fb) accept multiple device numbers by "<num>,<num>". * Show "Config: key=value" only if shortcut keys whose format is "proto:(echo)key=value" is pressed. * Merge patches: https://github.com/arakiken/mlterm/pull/5. https://github.com/arakiken/mlterm/pull/9. https://github.com/arakiken/mlterm/pull/12. https://github.com/arakiken/mlterm/pull/22. * Bug fixes: Fix https://github.com/arakiken/mlterm/issues/15. Fix https://github.com/arakiken/mlterm/issues/18. Fix https://github.com/arakiken/mlterm/issues/24. Fix https://github.com/arakiken/mlterm/issues/28. Fix failure of restarting mlterm on Android. (Enbugged at 3.9.1) Fix segfault if mlterm screen with -t option goes outside of the display. Fix emoji glyphs becoming too large with cairo. Fix corrupt value of "word_separator" option which mlconfig outputs. Fix incompatibility with xterm in pressing Shift + a-z keys etc if modifyOtherKeys is 1. (https://github.com/arakiken/mlterm/issues/21) Fix segfault in opening a new roxterm tab with libvte compatible library on wayland. - modified patches % etc.patch (refreshed) ==== python-Pillow ==== Version update (9.1.0 -> 9.1.1) Subpackages: python38-Pillow python38-Pillow-tk - update to 9.1.1 (bsc#1199713, CVE-2022-30595): * When reading past the end of a TGA scan line, reduce bytes left. CVE-2022-30595 * Do not open images with zero or negative height #6269 ==== python-matplotlib ==== Version update (3.5.1 -> 3.5.2) Subpackages: python38-matplotlib python38-matplotlib-tk - Skip a flaky leak test, move sigint test skip to all archs - Add matplotlib-pr22975-fixarray.patch, gh#matplotlib/matplotlib#22975, required for astropy 5.1 - update to 3.5.2: * Preliminary support for Windows on arm64 target has been added; this requires FreeType 2.11 or above. - drop matplotlib-pr22780-Pillow-deprecations.patch (upstream) ==== python-numpy ==== Version update (1.21.4 -> 1.21.6) - update to 1.21.6: * Do not forward `__(deep)copy__` calls of `_GenericAlias`... * Fix float16 einsum fastpaths using wrong tempvar * Print os error message when the executable not exist * Force ``npymath` ` to respect ``npy_longdouble`` * Fix failure to create aligned, empty structured dtype * provide a convenience function to replace npy_load_module * update wheel to version that supports python3.10 * Clear errors correctly in F2PY conversions * add a warningfilter to fix pytest workflow. * Help boost::python libraries at least not crash - drop bpo-45167-fixes.patch (upstream) ==== python-packaging ==== - Add patch to fix testsuite on big-endian targets + fix-big-endian-build.patch ==== rubygem-ruby-dbus ==== Version update (0.18.0.beta6 -> 0.18.0.beta7) - 0.18.0.beta7 API: * DBus.variant(type, value) is deprecated in favor of Data::Variant.new(value, member_type:) Bug fixes: * Client-side properties: When calling Properties.Set in ProxyObjectInterface#[]=, use the correct type (gh#mvidner/ruby-dbus#108). ==== sshfs ==== Version update (3.7.2 -> 3.7.3) - sshfs 3.7.3: * Minor bugfixes. * This is the last release from the current maintainer. SSHFS is now no longer maintained. - removed obsolete fix-typo-in-ssh_opts.patch ==== update-alternatives ==== Version update (1.21.1 -> 1.21.8) - version update to 1.21.8 * fix CVE-2022-1664 [bsc#1199944], dpkg -- security update * lot of changes, see changelog - modified patches % update-alternatives-suse.patch (refreshed) ==== vim ==== Version update (8.2.4877 -> 8.2.5038) Subpackages: gvim vim-data vim-data-common - Updated to version 8.2.5038, fixes the following problems * Valgrind warning for using uninitialized variable. * Screendump test may fail when using valgrind. * Vim9: misplaced elseif causes invalid memory access. * "P" in Visual mode still changes some registers. * Cannot make 'breakindent' use a specific column. * String interpolation only works in heredoc. * Test fails without the job/channel feature. (Dominique Pellé) * Test fails with the job/channel feature. * Vim9: redir in skipped block seen as assignment. * Channel log does not show invoking a timer callback. * Line number of lambda ignores line continuation. * Inconsistent capitalization in error messages. * Vim help presentation could be better. * Test failures because of changed error messages. * Distributed import files are not installed. * Buffer overflow with invalid command with composing chars. * Expression in command block does not look after NL when command is typed. * Comment inside an expression in lambda ignores the rest of the expression. * Coverity complains about pointer usage. * With latin1 encoding CTRL-W might go before the start of the command line. * Vim9 expression test fails without the job feature. * NULL pointer access when using invalid pattern. * Mouse wheel scrolling is inconsistent. * Cannot get the current cmdline completion type and position. * codecov includes MS-Windows install files. * codecov includes MS-Windows install header file. * Some users do not want a line comment always inserted. * No text formatting for // comment after a statement. * MODE_ enum entries names are too generic. * Imperfect coding. * The mode #defines are not clearly named. * Using execute() to define a lambda doesn't work. (Ernie Rael) * Popup_hide() does not always have effect. * String interpolation in :def function may fail. * Sometimes the cursor is in the wrong position. * Mouse in Insert mode test fails. * Fuzzy expansion of option names is not right. * Conceal character from matchadd() displayed too many times. * Can add invalid bytes with :spellgood. * Spell test fails because of new illegal byte check. * Mouse test fails on MS-Windows. * Test checks for terminal feature unnecessarily. * maparg() may return a string that cannot be reused. * Trailing backslash may cause reading past end of line. * #ifdef for crypt feature around too many lines. * Return type of remove() incorrect when using three arguments. * Various white space and cosmetic mistakes. * Off-by-one error in in statusline item. * Interpolated string expression requires escaping. * Crash with sequence of Perl commands. * Not easy to filter the output of maplist(). * A few more capitalization mistakes in error messages. * String interpolation fails when not evaluating. * With 'foldmethod' "indent" some lines are not included in the fold. (Oleg Koshovetc) * No test for what 8.2.4931 fixes. * Crash when matching buffer with invalid pattern. * matchfuzzypos() with "matchseq" does not have all positions. * Some code is never used. * '[ and '] marks may be wrong after undo. * Error when setting 'filetype' in help file again. * Changing 'switchbuf' may have no effect. * Text properties are wrong after "cc". (Axel Forsman) * Inconsistent use of white space. * Vim9: some code not covered by tests. * Text properties not adjusted when accepting spell suggestion. * Cannot use Perl heredoc in nested :def function. (Virginia Senioria) * Vim9: some code not covered by tests. * Text properties position wrong after shifting text. * Smart indenting done when not enabled. * GUI test will fail if color scheme changes. * With 'smartindent' inserting '}' after completion goes wrong. * Inserting line breaks text property spanning more then one line. * Text property in wrong position after auto-indent. * Reading past end of line with "gf" in Visual block mode. * Text properties in a wrong position after a block change. * A couple conditions are always true. * Using NULL regexp program. * Text properties that cross line boundary are not correctly updated for a deleted line. * Build error with a certain combination of features. * Files show up in git status. * Expanding path with "/**" may overrun end of buffer. * GUI: testing mouse move event depends on screen cell size. * Changing text in Visual mode may cause invalid memory access. * "eval 123" gives an error, "eval 'abc'" does not. * Vim9: interpolated string seen as range. * Vim9: compilation fails when using dict member when skipping. * Vim9: type error for list unpack mentions argument. * ":so" command may read after end of buffer. * Recursive command line loop may cause a crash. * Coverity complains about not restoring a saved value. * Memory access error when substitute expression changes window. * No error if engine selection atom is not at the start. * Accessing freed memory when line is flushed. * When 'shortmess' contains 'A' loading a session may still warn for an existing swap file. (Melker Österberg) * It is not possible to manipulate autocommands. * Colors in terminal window are not 100% correct. * Colors test fails in the GUI. * Dragging statusline fails for window with winbar. * PVS warns for possible array underrun. * Some github actions are outdated. * After deletion a small fold may be closable. * Textprop in wrong position when replacing multi-byte chars. * Cannot specify a function name for :defcompile. * Memory leak when :defcompile fails. * No test for hwat patch 8.1.0535 fixes. * Compiler warning for possibly uninitialized variable. (Tony Mechelynck) * smart/C/lisp indenting is optional, which makes the code more complex, while it only reduces the executable size a bit. * Tests are using legacy functions. * Still a compiler warning for possibly uninitialized variable. (Tony Mechelynck) * setbufline() may change Visual selection. (Qiming Zhao) * Python: changing hidden buffer can cause the display to be messed up. * Vim9: crash when using multiple funcref(). * Filetype test table is not properly sorted. * Checking translations affects the search pattern history. * deletebufline() may change Visual selection. * Cannot do bitwise shifts. * Right shift on negative number does not work as documented. * Compiler warning for uninitialized variable. (John Marriott) * Asan warns for undefined behavior. * Spell suggestion may use uninitialized memory. (Zdenek Dohnal) * When 'formatoptions' contains "/" wrongly wrapping a long trailing comment. * Fold may not be closeable after appending. * The terminal debugger uses various global variables. * Replacing an autocommand requires several lines. * Cannot select one character inside (). * After text formatting the cursor may be in an invalid position. * Byte offsets are wrong when using text properties. * Hoon and Moonscript files are not recognized. * Access before start of text with a put command. * Gcc 12.1 warns for uninitialized variable. * Vim9: some code is not covered by tests. * Cannot get the first screen column of a character. * Using 'imstatusfunc' and 'imactivatefunc' breaks 'foldopen'. * Build fails with normal features and +terminal. (Dominique Pellé) * 'completefunc'/'omnifunc' error does not end completion. * Substitute overwrites allocated buffer. * Using freed memory with "]d". * Vim9: a few lines not covered by tests. * Error for missing :endif when an exception was thrown. (Dani Dickstein) * Syntax regexp matching can be slow. * "textlock" is always zero. * autocmd_add() can only handle one event and pattern. * Cannot easily run the benchmarks. * Python 3 test fails without the GUI. * Build error with +eval but without +quickfix. Warning for uninitialized variable. * There is no way to get the byte index from a virtual column. * When splitting a window the changelist position moves. * Using two counters for timeout check in NFA engine. * Cursor position may be invalid after "0;" range. * A finished terminal in a popup window does not show a scrollbar. ==== virtualbox ==== Subpackages: virtualbox-guest-tools virtualbox-guest-x11 - Update fixes_for_kernel_5.18.patch to save and restore FPU status during interrupt. The change fixes boo#1199803. ==== virtualbox-kmp ==== - Update fixes_for_kernel_5.18.patch to save and restore FPU status during interrupt. The change fixes boo#1199803. ==== yast2-packager ==== Version update (4.5.4 -> 4.5.5) - Support managing system in a chroot (bsc#1199840) - 4.5.5

1 0

New Tumbleweed snapshot 20220518 released!
by Dominique Leuenberger 31 May '22

31 May '22

Please note that this mail was generated by a script. The described changes are computed based on the x86_64 DVD. The full online repo contains too many changes to be listed here. Please check the known defects of this snapshot before upgrading: https://openqa.opensuse.org/tests/overview?distri=opensuse&groupid=1&versio… Please do not reply to this email to report issues, rather file a bug on bugzilla.opensuse.org. For more information on filing bugs please see https://en.opensuse.org/openSUSE:Submitting_bug_reports Packages changed: NetworkManager (1.36.4 -> 1.38.0) autofs aws-cli (1.23.11 -> 1.24.1) cifs-utils (6.14 -> 6.15) kernel-firmware (20220509 -> 20220516) libdv libinput libxfce4ui libyui (4.4.0 -> 4.4.1) libyui-ncurses (4.4.0 -> 4.4.1) libyui-ncurses-pkg (4.4.0 -> 4.4.1) libyui-qt (4.4.0 -> 4.4.1) libyui-qt-graph (4.4.0 -> 4.4.1) libyui-qt-pkg (4.4.0 -> 4.4.1) mjpegtools okteta (0.26.7 -> 0.26.8) python-boto3 (1.22.11 -> 1.23.1) python-botocore (1.25.11 -> 1.26.1) transactional-update (4.0.0~rc2 -> 4.0.0~rc4) === Details === ==== NetworkManager ==== Version update (1.36.4 -> 1.38.0) Subpackages: NetworkManager-lang NetworkManager-pppoe libnm0 typelib-1_0-NM-1_0 - Update to version 1.38.0: + Add support for route type "throw". + Fix bug setting priority for IP addresses. + Static IPv6 addresses from "ipv6.addresses" are now preferred over addresses from DHCPv6, which are preferred over addresses from autoconf. This affects IPv6 source address selection, if the rules from RFC 6724, section 5 don't give a exhaustive match. + Static IPv6 addresses from "ipv6.addresses" are now interpreted with first address being preferred. Their order got inverted. This is now consistent with IPv4. + Wi-Fi hotspots will use a (stable) random channel number unless one is chosen manually. + Don't use unsupported SAE/WPA3 mode for AP mode. + NetworkManager will no longer advertise frequencies as supported when they're disallowed in configured regulatory domain. + Attempt to connect to WEP-encrypted Wi-Fi network will now fail gracefully with a recent version of wpa_supplicant when built without WEP support. As long as wpa_supplicant supports WEP, NetworkManager will continue to work. + Disable WPA3 transition mode for wifi.key-mgmt=wpa-psk if the NIC does not support PMF. This is known to cause problems in some setups. It is still possible to explicitly configure wifi.key-mgmt=sae for WPA3. + Add new dummy crypto backend "null" that does nothing. NetworkManager uses the crypto library when handling certificates for 802.1x profiles. + Veth devices with name "eth*" are now managed by default via the udev rule. This is to support managing the network in LXD containers. + The hostname received from DHCP is now shortened to the first dot (or to 64 characters, whatever comes first) if it's too long. + As the insecure WEP encryption for Wi-Fi network is phased out, nmcli now discourages its use when activating or modifying a profile. + Fix connectivity checks in case the check endpoint address resolves to multiple addresses. + Workaround libcurl blocking NetworkManager while resolving DNS names. + nmcli: indicate missing Wi-Fi hardware when showing rfkill setting. + nmcli: add connection migrate command to move a profile to a specified settings plugin. This allows to convert profiles in the deprecated ifcfg-rh format to keyfile. + Set "src" attribute for routes from DHCPv4 to the leased address. This helps with source address selection. + Various bugfixes and internal improvements. + Updated translations. - Recommend NetworkNanager-wifi from the main package: after the split, there is currently nothing pulling in NM-wifi. Preferably this would happen based on wifi chips prsence, but that is not yet done (boo#1199550). - Modify NetworkManager.spec: Split into a few small subpackages (bsc#1198128). ==== autofs ==== - Use OPTIONS instead of AUTOFS_OPTIONS in /etc/sysconfig/autofs (bsc#1199027) ==== aws-cli ==== Version update (1.23.11 -> 1.24.1) - Update to version 1.24.1 + For detailed changes see https://github.com/aws/aws-cli/blob/1.24.1/CHANGELOG.rst - Update Requires in spec file from setup.py ==== cifs-utils ==== Version update (6.14 -> 6.15) - Update to version 6.15 * CVE-2022-27239: mount.cifs: fix length check for ip option parsing Previous check was true whatever the length of the input string was, leading to a buffer overflow in the subsequent strcpy call. * mount.cifs: fix verbose messages on option parsing ==== kernel-firmware ==== Version update (20220509 -> 20220516) Subpackages: kernel-firmware-all kernel-firmware-amdgpu kernel-firmware-ath10k kernel-firmware-ath11k kernel-firmware-atheros kernel-firmware-bluetooth kernel-firmware-bnx2 kernel-firmware-brcm kernel-firmware-chelsio kernel-firmware-dpaa2 kernel-firmware-i915 kernel-firmware-intel kernel-firmware-iwlwifi kernel-firmware-liquidio kernel-firmware-marvell kernel-firmware-media kernel-firmware-mediatek kernel-firmware-mellanox kernel-firmware-mwifiex kernel-firmware-network kernel-firmware-nfp kernel-firmware-nvidia kernel-firmware-platform kernel-firmware-prestera kernel-firmware-qcom kernel-firmware-qlogic kernel-firmware-radeon kernel-firmware-realtek kernel-firmware-serial kernel-firmware-sound kernel-firmware-ti kernel-firmware-ueagle kernel-firmware-usb-network ucode-amd - Update to version 20220516 (git commit 251d29004ffc): * amdgpu: update beige goby firmware for 22.10 * amdgpu: update renoir firmware for 22.10 * amdgpu: update dimgrey cavefish firmware for 22.10 * amdgpu: update vega20 firmware for 22.10 * amdgpu: update yellow carp firmware for 22.10 * amdgpu: update vega12 firmware for 22.10 * amdgpu: update navy flounder firmware for 22.10 * amdgpu: update vega10 firmware for 22.10 * amdgpu: update raven2 firmware for 22.10 * amdgpu: update raven firmware for 22.10 * amdgpu: update sienna cichlid firmware for 22.10 * amdgpu: update green sardine firmware for 22.10 * amdgpu: update PCO firmware for 22.10 * amdgpu: update vangogh firmware for 22.10 * amdgpu: update navi14 firmware for 22.10 * amdgpu: update navi12 firmware for 22.10 * amdgpu: update navi10 firmware for 22.10 * amdgpu: update aldebaran firmware for 22.10 * linux-firmware: Update firmware file for Intel Bluetooth 9462 * linux-firmware: Update firmware file for Intel Bluetooth 9462 * linux-firmware: Update firmware file for Intel Bluetooth 9560 * linux-firmware: Update firmware file for Intel Bluetooth 9560 * linux-firmware: Update firmware file for Intel Bluetooth AX201 * linux-firmware: Update firmware file for Intel Bluetooth AX201 * linux-firmware: Update firmware file for Intel Bluetooth AX211 * linux-firmware: Update firmware file for Intel Bluetooth AX211 * linux-firmware: Update firmware file for Intel Bluetooth AX210 * linux-firmware: Update firmware file for Intel Bluetooth AX200 * linux-firmware: Update firmware file for Intel Bluetooth AX201 * mediatek: Update mt8192 SCP firmware ==== libdv ==== - Replace SDL-devel BuildRequires with pkgconfig(sdl): allow to use sdl12_compat as an alternative. ==== libinput ==== Subpackages: libinput-udev libinput10 - Enable building libinput-replay [boo#1190065] ==== libxfce4ui ==== Subpackages: libxfce4ui-2-0 libxfce4ui-lang libxfce4ui-tools typelib-1_0-Libxfce4ui-2_0 - Resolve rpmlint report "libxfce4ui-2-0.x86_64: E: shlib-policy-name-error SONAME: libxfce4kbd-private-3.so.0, expected package suffix: 3-0" - Move documentation to -devel package ==== libyui ==== Version update (4.4.0 -> 4.4.1) - Added a custom QTranslator for translations support for Qt Designer .ui files (bsc#1198097) - Renamed .ui files and toplevel classes in .ui files to conform to our naming standards (QY2*, YQ*) to avoid ambiguities with predefined Qt classes to work with our new custom QTranslator - Added TEXTDOMAIN file to support .ui files in y2makepot (@lslezak) - 4.4.1 ==== libyui-ncurses ==== Version update (4.4.0 -> 4.4.1) - Added a custom QTranslator for translations support for Qt Designer .ui files (bsc#1198097) - Renamed .ui files and toplevel classes in .ui files to conform to our naming standards (QY2*, YQ*) to avoid ambiguities with predefined Qt classes to work with our new custom QTranslator - Added TEXTDOMAIN file to support .ui files in y2makepot (@lslezak) - 4.4.1 ==== libyui-ncurses-pkg ==== Version update (4.4.0 -> 4.4.1) - Added a custom QTranslator for translations support for Qt Designer .ui files (bsc#1198097) - Renamed .ui files and toplevel classes in .ui files to conform to our naming standards (QY2*, YQ*) to avoid ambiguities with predefined Qt classes to work with our new custom QTranslator - Added TEXTDOMAIN file to support .ui files in y2makepot (@lslezak) - 4.4.1 ==== libyui-qt ==== Version update (4.4.0 -> 4.4.1) - Added a custom QTranslator for translations support for Qt Designer .ui files (bsc#1198097) - Renamed .ui files and toplevel classes in .ui files to conform to our naming standards (QY2*, YQ*) to avoid ambiguities with predefined Qt classes to work with our new custom QTranslator - Added TEXTDOMAIN file to support .ui files in y2makepot (@lslezak) - 4.4.1 ==== libyui-qt-graph ==== Version update (4.4.0 -> 4.4.1) - Added a custom QTranslator for translations support for Qt Designer .ui files (bsc#1198097) - Renamed .ui files and toplevel classes in .ui files to conform to our naming standards (QY2*, YQ*) to avoid ambiguities with predefined Qt classes to work with our new custom QTranslator - Added TEXTDOMAIN file to support .ui files in y2makepot (@lslezak) - 4.4.1 ==== libyui-qt-pkg ==== Version update (4.4.0 -> 4.4.1) - Added a custom QTranslator for translations support for Qt Designer .ui files (bsc#1198097) - Renamed .ui files and toplevel classes in .ui files to conform to our naming standards (QY2*, YQ*) to avoid ambiguities with predefined Qt classes to work with our new custom QTranslator - Added TEXTDOMAIN file to support .ui files in y2makepot (@lslezak) - 4.4.1 ==== mjpegtools ==== Subpackages: liblavfile-2_2-0 liblavjpeg-2_2-0 liblavplay-2_2-0 liblavrec-2_2-0 libmjpegutils-2_2-0 libmpeg2encpp-2_2-0 libmplex2-2_2-0 - mjpegtools's Makefiles try to use -lX11, but there is no BuildRequire for it - add it. ==== okteta ==== Version update (0.26.7 -> 0.26.8) Subpackages: libKasten4 libOkteta3 libkasten-lang libokteta-lang okteta-data okteta-lang - Update to 0.26.8 * Fixed: Float 32-bit editor in decoding table tool no longer rejects negative values & decimal places (kde#453819) * Fixed: PNG structure definition example now ensures use of big endianness (kde#452362) * Improved: translations * Improved: have more text vertically centered in lists in tool views * Changed: no more use of fixed font type for any list headers in tool views ==== python-boto3 ==== Version update (1.22.11 -> 1.23.1) - Update to version 1.23.1 * api-change:``resiliencehub``: [``botocore``] In this release, we are introducing support for Amazon Elastic Container Service, Amazon Route 53, AWS Elastic Disaster Recovery, AWS Backup in addition to the existing supported Services. This release also supports Terraform file input from S3 and scheduling daily assessments * api-change:``servicecatalog``: [``botocore``] Updated the descriptions for the ListAcceptedPortfolioShares API description and the PortfolioShareType parameters. * api-change:``discovery``: [``botocore``] Add Migration Evaluator Collector details to the GetDiscoverySummary API response * api-change:``sts``: [``botocore``] Documentation updates for AWS Security Token Service. * api-change:``workspaces-web``: [``botocore``] Amazon WorkSpaces Web now supports Administrator timeout control * api-change:``rekognition``: [``botocore``] Documentation updates for Amazon Rekognition. * api-change:``cloudfront``: [``botocore``] Introduced a new error (TooLongCSPInResponseHeadersPolicy) that is returned when the value of the Content-Security-Policy header in a response headers policy exceeds the maximum allowed length. - from version 1.23.0 * feature:Loaders: [``botocore``] Support for loading gzip compressed model files. * api-change:``grafana``: [``botocore``] This release adds APIs for creating and deleting API keys in an Amazon Managed Grafana workspace. - from version 1.22.13 * api-change:``ivschat``: [``botocore``] Documentation-only updates for IVS Chat API Reference. * api-change:``lambda``: [``botocore``] Lambda releases NodeJs 16 managed runtime to be available in all commercial regions. * api-change:``kendra``: [``botocore``] Amazon Kendra now provides a data source connector for Jira. For more information, see https://docs.aws.amazon.com/kendra/latest/dg/data-source-jira.html * api-change:``transfer``: [``botocore``] AWS Transfer Family now accepts ECDSA keys for server host keys * api-change:``iot``: [``botocore``] Documentation update for China region ListMetricValues for IoT * api-change:``workspaces``: [``botocore``] Increased the character limit of the login message from 600 to 850 characters. * api-change:``finspace-data``: [``botocore``] We've now deprecated CreateSnapshot permission for creating a data view, instead use CreateDataView permission. * api-change:``lightsail``: [``botocore``] This release adds support to include inactive database bundles in the response of the GetRelationalDatabaseBundles request. * api-change:``outposts``: [``botocore``] Documentation updates for AWS Outposts. * api-change:``ec2``: [``botocore``] This release introduces a target type Gateway Load Balancer Endpoint for mirrored traffic. Customers can now specify GatewayLoadBalancerEndpoint option during the creation of a traffic mirror target. * api-change:``ssm-incidents``: [``botocore``] Adding support for dynamic SSM Runbook parameter values. Updating validation pattern for engagements. Adding ConflictException to UpdateReplicationSet API contract. - from version 1.22.12 * api-change:``secretsmanager``: [``botocore``] Doc only update for Secrets Manager that fixes several customer-reported issues. * api-change:``ec2``: [``botocore``] This release updates AWS PrivateLink APIs to support IPv6 for PrivateLink Services and Endpoints of type 'Interface'. - Update BuildRequires and Requires from setup.py ==== python-botocore ==== Version update (1.25.11 -> 1.26.1) - Update to 1.26.1 * api-change:``resiliencehub``: In this release, we are introducing support for Amazon Elastic Container Service, Amazon Route 53, AWS Elastic Disaster Recovery, AWS Backup in addition to the existing supported Services. This release also supports Terraform file input from S3 and scheduling daily assessments * api-change:``servicecatalog``: Updated the descriptions for the ListAcceptedPortfolioShares API description and the PortfolioShareType parameters. * api-change:``discovery``: Add Migration Evaluator Collector details to the GetDiscoverySummary API response * api-change:``sts``: Documentation updates for AWS Security Token Service. * api-change:``workspaces-web``: Amazon WorkSpaces Web now supports Administrator timeout control * api-change:``rekognition``: Documentation updates for Amazon Rekognition. * api-change:``cloudfront``: Introduced a new error (TooLongCSPInResponseHeadersPolicy) that is returned when the value of the Content-Security-Policy header in a response headers policy exceeds the maximum allowed length. - from version 1.26.0 * feature:Loaders: Support for loading gzip compressed model files. * api-change:``grafana``: This release adds APIs for creating and deleting API keys in an Amazon Managed Grafana workspace. - from version 1.25.13 * api-change:``ivschat``: Documentation-only updates for IVS Chat API Reference. * api-change:``lambda``: Lambda releases NodeJs 16 managed runtime to be available in all commercial regions. * api-change:``kendra``: Amazon Kendra now provides a data source connector for Jira. For more information, see https://docs.aws.amazon.com/kendra/latest/dg/data-source-jira.html * api-change:``transfer``: AWS Transfer Family now accepts ECDSA keys for server host keys * api-change:``iot``: Documentation update for China region ListMetricValues for IoT * api-change:``workspaces``: Increased the character limit of the login message from 600 to 850 characters. * api-change:``finspace-data``: We've now deprecated CreateSnapshot permission for creating a data view, instead use CreateDataView permission. * api-change:``lightsail``: This release adds support to include inactive database bundles in the response of the GetRelationalDatabaseBundles request. * api-change:``outposts``: Documentation updates for AWS Outposts. * api-change:``ec2``: This release introduces a target type Gateway Load Balancer Endpoint for mirrored traffic. Customers can now specify GatewayLoadBalancerEndpoint option during the creation of a traffic mirror target. * api-change:``ssm-incidents``: Adding support for dynamic SSM Runbook parameter values. Updating validation pattern for engagements. Adding ConflictException to UpdateReplicationSet API contract. - from version 1.25.12 * api-change:``secretsmanager``: Doc only update for Secrets Manager that fixes several customer-reported issues. * api-change:``ec2``: This release updates AWS PrivateLink APIs to support IPv6 for PrivateLink Services and Endpoints of type 'Interface'. ==== transactional-update ==== Version update (4.0.0~rc2 -> 4.0.0~rc4) Subpackages: dracut-transactional-update libtukit4 transactional-update-zypp-config tukit tukitd - Version 4.0.0~rc4 - Fix building with GCC 12 - Fix stack overflow with very long commands / ids [bsc#1196149] - Use separate mount namespace for chroot, allowing overwriting the bind mounts from the update environment - this could have lead to data loss of the bind mount previously - Fix C error and exception handling for snapshots - Version 4.0.0~rc3 - Add Snapshot interface - Reworked signal handling: All public signals are sent from the main thread now, keeping the same sender for everything - Implement D-Bus call "Execute" for Transactions - Implement interface for listing Snapshots - Implement Reboot interface - Fix bug when using --continue on old snapshots - Fix hypothetical integer overflow in snapshot list [bsc#1196826] - Fix wrong sort order in status command [gh#openSUSE/transactional-update#80]

27 86

Re: opensuse server problems or outage?
by Joe Salmeri 30 May '22

30 May '22

On Mon, May 30, 2022 at 8:49 AM Christian Boltz <opensuse(a)cboltz.de> wrote: > > > Am Sonntag, 29. Mai 2022, 19:56:46 CEST schrieb Joe Salmeri: > > > > > https://download.opensuse.org/history/latest > > > > > > > > but if it fails it fails silently > > > > > > That's 1) - you sometimes get an empty file > > > > > > If you can reproduce this with "curl -v", I'd be interested to see > > > the output. > > > > It's pretty random when it happens. > > I made a copy of the /usr/bin/tumbleweed and changed the latest > > function to use curl -v > > > > Here's an example where latest was blank: > > > > PS /home/joe> ./tw2 status > > % Total % Received % Xferd Average Speed Time Time Time > > Current > > Dload Upload Total Spent Left > > Speed > > 0 0 0 0 0 0 0 0 --:--:-- --:--:-- > > --:--:-- 0* Trying 195.135.221.134:80... > > * Connected to download.opensuse.org (195.135.221.134) port 80 (#0) > > > > > GET /history/latest HTTP/1.1 > > > Host: download.opensuse.org > > > User-Agent: curl/7.83.1 > > > Accept: */* > > > > * Mark bundle as not supporting multiuse > > < HTTP/1.1 302 Found > > < Date: Sun, 29 May 2022 21:44:28 GMT > > < Server: Apache > > < content-length: 0 > > < location: http://mirrorcache-eu.opensuse.org/history/latest > > Ah, here we go. You get a HTTP redirect ("location: ..."), but curl > doesn't follow redirects automatically. > > That might be a side effect of our switch from mirrorbrain to mirrorcache > for http some weeks ago. (AFAIK mirrorbrain always directly served this > file.) (For https, we use mirrorbrain since about a year.) > > Adding -L to the curl call should help (it enables following the > "location:" header). > > Please also open a bugreport for tumbleweed-cli so that everybody gets > the -L added. > > > > That's 2) - download.o.o doesn't have latest Tumbleweed > > I have submitted the bug report on tumbleweed-cli. > > Yeah, it takes a little while after a build is published before the > > ISO shows up and tumbleweed status lists it as the latest, BUT, in > > this case a second build was published and neither build is showing > > up as available. That's why I reported the issue because it is > > definitely not normal for that to have occurred. > > Bernhard did manually call the publish script in the meantime (see > ticket) - that's not a fix, but at least a workaround. > [from #opensuse-project] > > Dominique has fixed the publishing issue.

1 0

dlopen on tumbleweed
by Christian Goll 30 May '22

30 May '22

Hello, the slurm package does not work on Tumbleweed any more, due to missing symbols. This is a Tumbleweed problem only and does not show up on Leap/SLES. If I compile slurm on the local machine, with the exact same configure and $CFLAGS everything works. So this must be a problem with the symbol stripping in osc/obs. I have opened a bug to track this issue https://bugzilla.suse.com/show_bug.cgi?id=1200030 Any thoughts, comments, tips or tricks to tackle this issue? kind regards, Christian -- Christian Goll CGoll(a)suse.com SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany Geschäftsführer: Ivo Totev, Andrew Myers, Andrew McDonald, Martje Boudien Moerman (HRB 36809, AG Nürnberg)

2 3

opensuse server problems or outage?
by Joe Salmeri 30 May '22

30 May '22

It appears to me that there might be a server problem and/or outage with opensuse.org. TumbleWeed build 20220526 was published yesterday and build 20220527 was published today as shown here https://openqa.opensuse.org/group_overview/1?limit_builds=20 however the latest build available here is 20220525 https://download.opensuse.org/tumbleweed/iso/ and this says the latest build is 20220525 https://download.opensuse.org/history/latest and the latest build in the list of builds is 20220525 When a new build is published it takes a little while for the ISO to appear and for the build to appear in the history location but it is not normal for a 2nd build to be published and the 1st build has still not appeared Could someone please look into this? I sent a message to the admin too.

8 10

Suggestion: ftrivial-auto-var-init=whatever by default
by Cristian Rodríguez 30 May '22

30 May '22

Hi: Bugs due to the use of uninitialized memory are rampant, not only in the kernel where it is extremely dangerous but it is even more widespread in userspace code. I suggest to: Use -ftrivial-auto-var-init=pattern during LEAP development stage, and -ftrivial-auto-var-init=zero in release builds and tumbleweed or some different flag combination depending on which project we are.. staging vs published trees or somethin.. If there is any heinous performance or code size regression on any component this could be disabled on a per-package basis or use__attribute__((uninitialized)) in the offending code. Cheers.

3 2

New Tumbleweed snapshot 20220527 released!
by Dominique Leuenberger 30 May '22

30 May '22

Please note that this mail was generated by a script. The described changes are computed based on the x86_64 DVD. The full online repo contains too many changes to be listed here. Please check the known defects of this snapshot before upgrading: https://openqa.opensuse.org/tests/overview?distri=opensuse&groupid=1&versio… Please do not reply to this email to report issues, rather file a bug on bugzilla.opensuse.org. For more information on filing bugs please see https://en.opensuse.org/openSUSE:Submitting_bug_reports Packages changed: amavisd-new automake binutils grep ppp qemu === Details === ==== amavisd-new ==== Subpackages: amavisd-new-docs - amavis fails to start (bsc#1199755) Add missed requirement - Remove server:mail/amavisd-new/amavis.service since it results Assignment outside of section warning ==== automake ==== - update automake.keyring: use release-team keyring - don't reference source URL when the linked sources change over time ==== binutils ==== Subpackages: libctf-nobfd0 libctf0 - Use https for variosu links. - Update binutils-2.38-branch.diff.gz (to 93054037f1e304e) in order to include PR29087. - Enable multitarget build on riscv64 - On SLE15 and later, use make -Oline to synchronize configure output by lines ==== grep ==== Subpackages: grep-lang - use release keyring rather than full one for validation - Do not link an unversioned file by URL (and refresh keyring) ==== ppp ==== - Add ppp-compiling-with-clang-encounters-an-error-in-eap-tls..patch that fixed the following rpmlint error: executable-stack (Badness: 10000) /usr/sbin/pppd ==== qemu ==== Subpackages: qemu-accel-tcg-x86 qemu-audio-spice qemu-block-curl qemu-block-rbd qemu-chardev-spice qemu-guest-agent qemu-hw-display-qxl qemu-hw-display-virtio-gpu qemu-hw-display-virtio-gpu-pci qemu-hw-display-virtio-vga qemu-hw-usb-host qemu-hw-usb-redirect qemu-hw-usb-smartcard qemu-ipxe qemu-ksm qemu-kvm qemu-lang qemu-microvm qemu-seabios qemu-sgabios qemu-tools qemu-ui-curses qemu-ui-gtk qemu-ui-opengl qemu-ui-spice-app qemu-ui-spice-core qemu-vgabios qemu-x86 - It has been observed that building QEMU with _FORTIFY_SOURCE=3 causes problem (see bsc#1199924). Force it to =2 for now, while we investigate the issue. - Backport a GCC 12 aarch64 build fix (bsc#1199625) * Patches added: block-qdict-Fix-Werror-maybe-uninitializ.patch

1 0

Something serious seems wrong with installation system
by Dave Plater 30 May '22

30 May '22

I changed my mother board and had to convert a legacy boot to efi and messed it up so I attempted to install Leap:15.4 fresh. For this I used a downloaded image installed on a usb stick first via dd_rescue direct and the last Tumbleweed stick via live-fat-stick. All of these attempts have had "wrong digest" errors which I ignored and eventually crashed. I thought this was due to my new mother board maybe so I decided to use a spare known to be good hard drive to install via my working Tumbleweed laptop (it already uses efi) using a usb adapter. Still "wrong digest" errors and they appear to be exactly the same files. Help!!! I have important work to finish and will be bankrupt if I don't The one thing related to this pc is that I burned the usb stick on this pc (new motherboard and intermittent crashes) but the micro sd card in reader that I used on my first attempt to install with was burned on the laptop. The packages that fail appear to be the same ones every time. I found a similar problem on the openSUSE forum but it apparently just went away for an unknown reason. PC = generation 4 msi H81I motherboard, 8 gig ram and i5 cpu with new hard disk (my old one started giving smart errors so I replaced it with a new one.) Laptop 6th generation HP Pavillion with i3 6006u cpu and 4 gig ram Please can somebody shed some light on my dilemma. See also https://bugzilla.suse.com/show_bug.cgi?id=1199903 for the one package that stubbornly refuses to install Thanks Dave Plater

2 2