June 2018 - openSUSE Factory - openSUSE Mailing Lists

[opensuse-factory] IDEA: NetworkManager by default for Desktops
by Richard Brown 05 Jul '18

05 Jul '18

Hi All, In both Leap & Tumbleweed we currently have the following logic for choosing which tool manages your network after installation "IF laptop AND NetworkManager is being installed THEN use NetworkManager, ELSE wicked" I've noticed an increasing trend that this can lead to confusion. Many users from other distributions expect NetworkManager on their desktops regardless of whether they are using a laptop or a desktop. GNOME expects NetworkManager by default and shows a rather unpleasant error when loading up the settings screen without it.[1] There are certainly classes of hardware where NetworkManager might be wanted that will never match "IF laptop" - eg. my personal Inter Compute Stick. And "IF laptop" is dependant on YaST correctly detecting you're using a laptop, which isn't 100% accurate, so occasionally even laptops might end up with wicked unexpectedly. I would like to propose a possible solution. It looks like we should be able to pin the choice of network tool to specific system role. I would like to propose that the installation options KDE and GNOME therefore always have NetworkManager by default Server & Transactional will always have wicked by default Custom will keep the current "autodetect" behaviour, because we cant be sure that the environment that the user is installing has support for NetworkManager in this case. Whatever is setup "by default" during installation will be able to be changed in YAST after the installation, just like it is today. What do we all think? If there is strong support for this idea I think Ludwig is prepared to let me try and slip it in as a late feature for Leap 15, so please let your opinions be heard loudly and quickly either way. Regards, Richard [1] http://paste.opensuse.org/33301696 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

42 131

[opensuse-factory] NGINX Ingress Controller for Kubernetes - new packages
by Michal Rostecki 03 Jul '18

03 Jul '18

Hi, I recently submitted 17 packages to Factory and I would like to explain the purpose of those requests. Kubernetes has a concept of Ingress[0] which allows to expose a service running inside cluster to the external network. Although to be able to use that functionality, user needs to deploy the Ingress Controller[1]. The most popular, open source and upstream supported implementation is NGINX Ingress Controller[2]. I would like to bring the support of Ingress in Kubic and SUSE CaaSP projects, which means that we need a container image and packages for every component needed by NGINX Ingress Controller. That's why I started with submitting all needed packages to Factory. This is the list of submit requests I made: https://build.opensuse.org/request/show/617519 https://build.opensuse.org/request/show/617660 https://build.opensuse.org/request/show/617664 https://build.opensuse.org/request/show/617665 https://build.opensuse.org/request/show/617667 https://build.opensuse.org/request/show/617668 https://build.opensuse.org/request/show/617669 https://build.opensuse.org/request/show/617670 https://build.opensuse.org/request/show/617671 https://build.opensuse.org/request/show/617672 https://build.opensuse.org/request/show/617673 https://build.opensuse.org/request/show/617731 https://build.opensuse.org/request/show/617734 https://build.opensuse.org/request/show/617745 https://build.opensuse.org/request/show/617750 https://build.opensuse.org/request/show/617752 https://build.opensuse.org/request/show/617753 I would be very thankful for feedback and reviews. Cheers, Michal [0] https://kubernetes.io/docs/concepts/services-networking/ingress/ [1] https://kubernetes.io/docs/concepts/services-networking/ingress/#ingress-co… [2] https://github.com/kubernetes/ingress-nginx -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

3 3

[opensuse-factory] devel:languages:python diet reduction
by Tomas Chvatal 03 Jul '18

03 Jul '18

Hello all, We will have to split a bit the d:l:py due to its monstrous size (2700 packages) where we are unable to verify the stack actually kinda works before Tumbleweed integration. We will split it based on common purpose and use d:l:py as base: devel:languages:python:django devel:languages:python:azure devel:languages:python:aws ... If you happen to have some packages or interest in python modules and think of some connection between couple of them please drop me an email for subproject creation and I will happily create it for you and set it up for TW integration. As a bit additional tweak I would move all of the non-Tumbleweed integrated stuff in to different project. At this moment it is full of various linked packages and things that were only added to develprj without being submitted to Tumbleweed. This prompted people to add this project on their released openSUSE versions (leading to quite interestingly broken systems if they do zypper dup with vendor change)... The change will be done by creating special project d:l:py:misc that would have building enabled but publishing disabled. This will hopefully motivate people to properly integrate the packages while not losing all the work that was done on them so far. Cheers Tom [1] https://build.opensuse.org/project/show/devel:languages:python

6 11

[opensuse-factory] Upcoming changes to HiDPI scaling in Qt
by Fabian Vogt 03 Jul '18

03 Jul '18

Hi, tl;dr section at the bottom. During Leap 15.0 development, a focus was on the experience with HiDPI (>96 pixels per inch) monitors. Some bugs got reported which indicated that there were some issues especially in mixed-DPI environments. While they generally work fine after some (extensive) manual configuration, the automatic configuration wasn't quite that good. Fixing this will also improve the experience during the installation (no manual configuration possible) and on desktops which don't have a UI to configure it. X was not designed for HiDPI and especially not mixed-DPI environments. It has only a single dpi specifier for the whole screen (spanning all monitors) and it is on all drivers, except the proprietary Nvidia driver, 96x96dpi. Each monitor geometry is available via XRandR though. Now Qt comes into play: It uses the screen DPI (so 96dpi) as global DPI and applies per-monitor scale factors on top. There is a font scale factor which is computed by (screen DPI)/(primary monitor DPI). If your primary monitor has 96dpi and your secondary 192dpi, it results in a 1x font scale and 1x/2x scale factors, as expected. If you set the 192dpi monitor as primary though, it gets ugly: The font scale is now 96/192 = 0.5x and 1x/2x scale factors mean that applications on either monitor are half of the expected size (1x on the primary and 0.5x on the secondary). This is reported upstream as https://bugreports.qt.io/browse/QTBUG-67928 The solution: Ignore the 96dpi which X reports by default and only use the monitor geometry as scale. The good: The font size on HiDPI and especially mixed-DPI environments is now scaled by default. The bad: You might unintentionally rely on the 96dpi X reports if you have a HiDPI monitor but keep applications unscaled. That requires manual configuration now. The ugly: If you have a primary monitor with != 96dpi, this is a behaviour change and results in changed appearance of applications. YaST for example uses bigger fonts, but the UI elements stay the same size. tl;dr: Look out for changes in application appearance and report any bugs. If you have a HiDPI system with custom configuration for DPI or scale, try to reset it to the defaults again and check whether it behaves as expected. You can revert to the old behaviour by setting "Xft.dpi: 96" (most configuration dialogs call it "Force font DPI") or by setting QT_FONT_DPI=96 in the environment. Cheers, Fabian -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

5 12

[opensuse-factory] Opening private bugs
by Simon Lees 02 Jul '18

02 Jul '18

Hi all, As many people are aware the sharing of code between Leap and SLE combined with SUSE's "Factory First" policy has often lead to the need for openSUSE community members needing to read bugs that have been marked as private because they were originally created for SLE. For some time SUSE has been happy for SUSE employee's to review and disclose the contents of private bugs as long as we don't share any customer information. We can either do this by making any such confidential info private in the bug should it exist then making the bug public or in other cases providing a summary of the bug and discussion, even if in some cases all we can say is "this is a private customer feature" In order to make this process slightly easier and more formal during las weeks board discussion the board decided to the mailing list opensuse-bugshare(a)opensuse.org where community members can send an email then a volunteer SUSE employee can read the bug report then either open it or provide a summary. Currently this is a short term solution, SUSE Engineering understands this problem and will be working toward a better solution longer term. If you are a SUSE employee and you are happy to help with this process shoot me an email and i'll get you subscribed to the list. Cheers -- Simon Lees (Simotek) http://simotek.net Emergency Update Team keybase.io/simotek SUSE Linux Adelaide Australia, UTC+10:30 GPG Fingerprint: 5B87 DB9D 88DC F606 E489 CEC5 0922 C246 02F0 014B

18 70

[opensuse-factory] Howto fix: Sources could not be expanded: conflict in file firefox-esr.spec in branched package
by David C. Rankin 01 Jul '18

01 Jul '18

All, Need a little help with build service. I branched firefox-esr to my build service to build 52.8.1 which built fine originally. I went back today to update sources and spec for 52.9.0 and now build service is telling me: Sources could not be expanded: conflict in file firefox-esr.spec Why is there a conflict? Of course I had to update the .spec to change versions and sources for firefox-52.8.1-source.tar.xz l10n-52.8.1.tar.xz and now firefox-52.9.0-source.tar.xz l10n-52.9.0.tar.xz How do I tell it "Yes, these are the changes I want in my repo?" https://build.opensuse.org/package/show/home:drankinatty:branches:mozilla/f… Do I need to delete this branch and re-branch firefox-esr again? -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

3 5

[opensuse-factory] New Tumbleweed snapshot 20180628 released!
by Dominique Leuenberger 30 Jun '18

30 Jun '18

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… When you reply to report some issues, make sure to change the subject. It is not helpful to keep the release announcement subject in a thread while discussing a specific problem. Packages changed: bcm43xx-firmware gnome-builder (3.28.2 -> 3.28.3) gstreamer-plugins-bad gstreamer-plugins-good kernel-source (4.17.2 -> 4.17.3) mpg123 openafs perl-Cpanel-JSON-XS (4.02 -> 4.04) units (2.16 -> 2.17) v4l2loopback (0.11.0_k4.17.2_1 -> 0.11.0_k4.17.3_1) vala (0.40.6 -> 0.40.7) xdg-desktop-portal-kde (5.13.1 -> 5.13.2) yast2-journal (4.1.0 -> 4.1.1) yast2-tftp-server (4.1.0 -> 4.1.1) === Details === ==== bcm43xx-firmware ==== - Add brcmfmac4356-pcie.txt for BCM 4356 PCI (bsc#1099149) ==== gnome-builder ==== Version update (3.28.2 -> 3.28.3) Subpackages: gnome-builder-doc gnome-builder-lang gnome-builder-plugin-jedi gnome-builder-plugin-jhbuild gnome-builder-plugin-vala-pack - Update to version 3.28.3: + Fixes for a number of crashers and potential data loss. + The editor more reliably restores cursor position. + The todo plugin more aggresively ignores some common build tooling directories. + Updated translations. ==== gstreamer-plugins-bad ==== Subpackages: gstreamer-plugins-bad-lang libgstadaptivedemux-1_0-0 libgstbadaudio-1_0-0 libgstbadvideo-1_0-0 libgstbasecamerabinsrc-1_0-0 libgstcodecparsers-1_0-0 libgstisoff-1_0-0 libgstmpegts-1_0-0 libgstphotography-1_0-0 libgsturidownloader-1_0-0 libgstwayland-1_0-0 libgstwebrtc-1_0-0 - Conditionalize pkgconfig(libsrtp2) and pkgconfig(nice) BuildRequires: fix build for Leap 42.3. ==== gstreamer-plugins-good ==== Subpackages: gstreamer-plugins-good-extra gstreamer-plugins-good-jack gstreamer-plugins-good-lang - Conditionalize pkgconfig(gtk+-wayland-3.0) BuildRequires: fix build for Leap 42.3. ==== kernel-source ==== Version update (4.17.2 -> 4.17.3) Subpackages: kernel-default kernel-default-devel kernel-devel kernel-docs kernel-macros kernel-syms - Linux 4.17.3 (bnc#1012628). - net: aquantia: fix unsigned numvecs comparison with less than zero (bnc#1012628). - bonding: re-evaluate force_primary when the primary slave name changes (bnc#1012628). - cdc_ncm: avoid padding beyond end of skb (bnc#1012628). - ipv6: allow PMTU exceptions to local routes (bnc#1012628). - net: dsa: add error handling for pskb_trim_rcsum (bnc#1012628). - net: phy: dp83822: use BMCR_ANENABLE instead of BMSR_ANEGCAPABLE for DP83620 (bnc#1012628). - net/sched: act_simple: fix parsing of TCA_DEF_DATA (bnc#1012628). - tcp: verify the checksum of the first data segment in a new connection (bnc#1012628). - tls: fix use-after-free in tls_push_record (bnc#1012628). - tls: fix waitall behavior in tls_sw_recvmsg (bnc#1012628). - udp: fix rx queue len reported by diag and proc interface (bnc#1012628). - net: in virtio_net_hdr only add VLAN_HLEN to csum_start if payload holds vlan (bnc#1012628). - hv_netvsc: Fix a network regression after ifdown/ifup (bnc#1012628). - ext4: fix hole length detection in ext4_ind_map_blocks() (bnc#1012628). - ext4: update mtime in ext4_punch_hole even if no blocks are released (bnc#1012628). - ext4: do not allow external inodes for inline data (bnc#1012628). - ext4: bubble errors from ext4_find_inline_data_nolock() up to ext4_iget() (bnc#1012628). - ext4: correctly handle a zero-length xattr with a non-zero e_value_offs (bnc#1012628). - ext4: fix fencepost error in check for inode count overflow during resize (bnc#1012628). - driver core: Don't ignore class_dir_create_and_add() failure (bnc#1012628). - Btrfs: allow empty subvol= again (bnc#1012628). - Btrfs: fix clone vs chattr NODATASUM race (bnc#1012628). - Btrfs: fix memory and mount leak in btrfs_ioctl_rm_dev_v2() (bnc#1012628). - btrfs: return error value if create_io_em failed in cow_file_range (bnc#1012628). - btrfs: scrub: Don't use inode pages for device replace (bnc#1012628). - ALSA: usb-audio: Disable the quirk for Nura headset (bnc#1012628). - ALSA: hda/realtek - Enable mic-mute hotkey for several Lenovo AIOs (bnc#1012628). - ALSA: hda/conexant - Add fixup for HP Z2 G4 workstation (bnc#1012628). - ALSA: hda - Handle kzalloc() failure in snd_hda_attach_pcm_stream() (bnc#1012628). - ALSA: hda: add dock and led support for HP EliteBook 830 G5 (bnc#1012628). - ALSA: hda: add dock and led support for HP ProBook 640 G4 (bnc#1012628). - x86/MCE: Fix stack out-of-bounds write in mce-inject.c: Flags_read() (bnc#1012628). - smb3: fix various xid leaks (bnc#1012628). - smb3: on reconnect set PreviousSessionId field (bnc#1012628). - CIFS: 511c54a2f69195b28afb9dd119f03787b1625bb4 adds a check for session expiry (bnc#1012628). - cifs: For SMB2 security informaion query, check for minimum sized security descriptor instead of sizeof FileAllInformation class (bnc#1012628). - nbd: fix nbd device deletion (bnc#1012628). - nbd: update size when connected (bnc#1012628). - nbd: use bd_set_size when updating disk size (bnc#1012628). - blk-mq: reinit q->tag_set_list entry only after grace period (bnc#1012628). - bdi: Move cgroup bdi_writeback to a dedicated low concurrency workqueue (bnc#1012628). - cpufreq: Fix new policy initialization during limits updates via sysfs (bnc#1012628). - cpufreq: ti-cpufreq: Fix an incorrect error return value (bnc#1012628). - cpufreq: governors: Fix long idle detection logic in load calculation (bnc#1012628). - libata: zpodd: small read overflow in eject_tray() (bnc#1012628). - libata: Drop SanDisk SD7UB3Q*G1001 NOLPM quirk (bnc#1012628). - nvme/pci: Sync controller reset for AER slot_reset (bnc#1012628). - w1: mxc_w1: Enable clock before calling clk_get_rate() on it (bnc#1012628). - x86/vector: Fix the args of vector_alloc tracepoint (bnc#1012628). - x86/apic/vector: Prevent hlist corruption and leaks (bnc#1012628). - x86/apic: Provide apic_ack_irq() (bnc#1012628). - x86/ioapic: Use apic_ack_irq() (bnc#1012628). - x86/platform/uv: Use apic_ack_irq() (bnc#1012628). - irq_remapping: Use apic_ack_irq() (bnc#1012628). - genirq/generic_pending: Do not lose pending affinity update (bnc#1012628). - genirq/affinity: Defer affinity setting if irq chip is busy (bnc#1012628). - genirq/migration: Avoid out of line call if pending is not set (bnc#1012628). - x86/intel_rdt: Enable CMT and MBM on new Skylake stepping (bnc#1012628). - media: uvcvideo: Prevent setting unavailable flags (bnc#1012628). - media: rc: ensure input/lirc device can be opened after register (bnc#1012628). - iwlwifi: fw: harden page loading code (bnc#1012628). - orangefs: set i_size on new symlink (bnc#1012628). - orangefs: report attributes_mask and attributes for statx (bnc#1012628). - HID: intel_ish-hid: ipc: register more pm callbacks to support hibernation (bnc#1012628). - HID: wacom: Correct logical maximum Y for 2nd-gen Intuos Pro large (bnc#1012628). - fs/binfmt_misc.c: do not allow offset overflow (bnc#1012628). - mm, page_alloc: do not break __GFP_THISNODE by zonelist reset (bnc#1012628). - commit e8dc1b5 - platform/x86: ideapad-laptop: Apply no_hw_rfkill to Y20-15IKBM, too (bsc#1098626). - commit cbe83fb - Input: psmouse - fix button reporting for basic protocols (bnc#1098392). - commit 7faa585 - vhost: fix info leak due to uninitialized memory (bsc#1092472 CVE-2018-1118). - commit 7372950 - x86/stacktrace: Do not unwind after user regs (bnc#1058115). - x86/stacktrace: Remove STACKTRACE_DUMP_ONCE (bnc#1058115). - x86/stacktrace: Clarify the reliable success paths (bnc#1058115). - x86/stacktrace: Do not fail for ORC with regs on stack (bnc#1058115). - x86/unwind/orc: Detect the end of the stack (bnc#1058115). - x86/stacktrace: Enable HAVE_RELIABLE_STACKTRACE for the ORC unwinder (bnc#1058115). - Delete patches.suse/0001-x86-stacktrace-do-now-unwind-after-user-regs.patch. - Delete patches.suse/0002-x86-stacktrace-make-clear-the-success-paths.patch. - Delete patches.suse/0003-x86-stacktrace-remove-STACKTRACE_DUMP_ONCE-from-__sa.patch. - Delete patches.suse/0004-x86-stacktrace-do-not-fail-for-ORC-with-regs-on-stac.patch. - Delete patches.suse/0005-x86-stacktrace-orc-mark-it-as-reliable.patch. Replace the ORC patches by the upstream patches. Finally... - commit 33a2d86 - s390: Correct register corruption in critical section cleanup (boo#1095717). - commit 4594a9e - ACPICA: AML parser: attempt to continue loading table after error (bsc#1098074). - commit fc39d10 - macros.kernel-source: define linux_arch for KMPs (boo#1098050). CONFIG_64BIT is no longer defined so KMP spec files need to include %{?linux_make_arch} in any make call to build modules or descent into the kernel directory for any reason. - commit 5dc40af - rpm: ignore CONFIG_GCC_VERSION when checking for oldconfig changes Since 4.18-rc1, "make oldconfig" writes gcc version and capabilities into generated .config. Thus whenever we build the package or run checks with different gcc version than used to update config/*/*, check for "outdated configs" fails. As a quick band-aid, omit the lines with CONFIG_GCC_VERSION from both configs before comparing them. This way, the check won't fail unless run with newer gcc which would add new capabilities. More robust solution will require a wider discussion. - commit 546ef32 ==== mpg123 ==== Subpackages: libmpg123-0 mpg123-openal mpg123-pulse - Conditionalize pkgconfig(openal) BuildRequires and mpg123-openal sub-package, fix build for SLE12 SP3. ==== openafs ==== Subpackages: openafs-client openafs-kmp-default - adjust building of KMP to new kernels (see boo 1098050) add patch add_arch_to_linux_kernel_make.patch for this - add libtirpc-devel to BuildRequires: - minor cleanups ==== perl-Cpanel-JSON-XS ==== Version update (4.02 -> 4.04) - updated to 4.04 see /usr/share/doc/packages/perl-Cpanel-JSON-XS/Changes 4.04 2018-06-22 (rurban) - Fix bignum NaN/inf handling (#78 reported by Slaven Rezic) - Move author tests to xt/ as suggested in #106, added a make xtest target. Fixes a test fail with ASAN. - updated to 4.03 see /usr/share/doc/packages/perl-Cpanel-JSON-XS/Changes 4.03 2018-06-21 (rurban) - Add sereal cpanel_json_xs type (#110 James Rouzier) - Fix bencode/bdecode methods in cpanel_json_xs (#111 Fulvio Scapin) - Overload ne operator for JSON::PP::Boolean (#107 tevfik1903) - Add a missing semicolon to a documentation example (#104 E. Choroba) ==== units ==== Version update (2.16 -> 2.17) - units 2.17: * Updates to units database * Update units_cur to handle bad records returned by yahoo server * Move currency.units out of /usr/share because the file is not static and /usr/share is supposed to house static files. The new location is $sharedstatedir which may be /var/lib ==== v4l2loopback ==== Version update (0.11.0_k4.17.2_1 -> 0.11.0_k4.17.3_1) - Use %{?linux_make_arch} when building kernel modules (boo#1098050). ==== vala ==== Version update (0.40.6 -> 0.40.7) Subpackages: libvala-0_40-0 - Update to version 0.40.7: + Various improvements and bug fixes: - Update bugtracker links to https://gitlab.gnome.org/GNOME/vala/issues. - codegen: . Handle delegate_target attribute of fields. . Free generic elements of glib collections. . Fix warning for source_funcs parameter of g_source_new(). . Actually treat GLib.Source as compact class. . Custom abstract methods of GLib.Source are handled differently. . Consistently use gpointer for delegate targets. . Unify some delegate-type check patterns. . Use default_init instead of base_init when registering interfaces. - girparser: Add support for bool delegate_target. - girwriter: . Output deprecated="1". . Construct-only properties don't have a setter method. . Don't write any custom attributes. - valadoc: Fix TreeBuilder.create_array() for stacked Arrays. + Bindings: - glib-2.0: . Do not mark simple-type out-parameters as nullable. . Make Source.attach() use the default MainContext by default. . Make Bytes.slice() use the memory-efficient Bytes.from_bytes(). . Fix Bytes.with_free_func(). - gio-2.0: . Mark ActionEntry.parameter_type/state fields as nullable. . Delegate fields of DBus*VTable do not have implicit target fields. . Apply delegate_target = false on ActionEntry callbacks and make them weak. - Add default to all io_priority parameters. - gstreamer: Cherry-pick some fixes from 0.42, add gst-editing-services-1.0. - gtk+-3.0: Update to 3.22.30+9ac7f906. - gtk+-4.0: Update to 3.93.0+29fee2e8. - webkit2gtk-4.0: Update to 2.21.4. - Update GIR-based bindings. ==== xdg-desktop-portal-kde ==== Version update (5.13.1 -> 5.13.2) Subpackages: xdg-desktop-portal-kde-lang - Update to 5.13.2 * New bugfix release * For more details please see: * https://www.kde.org/announcements/plasma-5.13.2.php - Changes since 5.13.1: * Stop streaming when session is closed and allow some stream negotiation ==== yast2-journal ==== Version update (4.1.0 -> 4.1.1) - Change namespace according to YaST naming conventions. Now this package uses Y2Journal (needed for fate#319428). - 4.1.1 ==== yast2-tftp-server ==== Version update (4.1.0 -> 4.1.1) - Adapt code to use new namespace Y2Journal (needed for fate#319428). - 4.1.1 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

1 0

[opensuse-factory] Tumbleweed - Review of the weeks 2018/24-26
by Dominique Leuenberger / DimStar 29 Jun '18

29 Jun '18

Dear Tumbleweed users and hackers, It’s been 3 weeks since my last review – Vacation treated me well and I was actually managing to stay away from computers for the whole time (not the mobile though… maybe next time). In my last review, I explain about all the trouble we had with the snapshots, but the last one I prepared for building, 0606, was finally published (so I at least hander ‘something working’ over to Max. I’m sure he was happy about that). As Tumbleweed had to be handled ‘beside the regular tasks without a dedicated release manager’, it was understandably having a slower release cadence. In total, during the last 3 weeks, there were 10 snapshots published to the users: 0606, 0613, 0615, 0618, 0620, 0622, 0623, 0625, 0626 and 0627, which, in summary, contained those updates: * GCC 8.1.1 is used to build the entire distribution * Qt 5.11.0 & 5.11.1 * Mesa 18.1.1 & 18.1.2 * KDE Plasma 5.13.0 & 5.13.1 * KDE Applications 18.04.2 * Linux kernel 4.17.1 and 4.17.2 * KDE Frameworks 5.47.0 * FFmpeg 4.0 (FFmpeg 3.x is still available in the main repo, but should be phased out) * PulseAudio 12.0 * git 2.18.0 In the last days, we changed the setup of the rings and staging projects for openSUSE:Factory and simplified the things a bit. From now on, we only manage Ring0 (minimal bootstrap) and ring1 (things we build pre=integration, guaranteed to always build). Everything that used to be ring2 before (e.g. KDE, GNOME) has been merged into ring1. For you as a contributor that means no direct change (except less confusion when looking for failures, as Staging:X now is all there is, no more Staging:X:DVD to know about and find), but the tooling used could be drastically simplified, which was worth this effort. All Stagings needed to be reset for this though, which means some things might take a tad longer than they normally would. The staged changes currently are: * linux-glibc-devel 4.17 (one more failure, opal, which I submitted a fix to the devel project today) * X.org 1.12.0 * acpica 20180508: breaks xen and VirtualBox * glibc: drop support for bundle-lang-* packages (they have been eliminated already a while ago in Tumbleweed * fwupdate 11: support for Lenovo machines * KDE Plasma 5.13.2 * Mozilla Firefox 61 Cheers, and happy weekend, Dominique

2 1

[opensuse-factory] New Tumbleweed snapshot 20180627 released!
by Dominique Leuenberger 29 Jun '18

29 Jun '18

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… When you reply to report some issues, make sure to change the subject. It is not helpful to keep the release announcement subject in a thread while discussing a specific problem. Packages changed: Mesa (18.1.1 -> 18.1.2) Mesa-drivers (18.1.1 -> 18.1.2) NetworkManager-applet (1.8.12 -> 1.8.14) bison (3.0.4 -> 3.0.5) discount (2.2.2 -> 2.2.3a) dracut ethtool (4.16 -> 4.17) gcc8 (8.1.1+r260570 -> 8.1.1+r261583) git (2.17.1 -> 2.18.0) gstreamer-rtsp-server gstreamer-transcoder (1.12.2 -> 1.14.0) hplip installation-images-Kubic (14.375 -> 14.376) kcm_tablet (3.0.0 -> 3.1.0) krb5 libtirpc libzypp (17.3.1 -> 17.4.0) mailman (2.1.26 -> 2.1.27) obs-service-source_validator (0.17 -> 0.18) perl-DateTime-Locale (1.200000 -> 1.220000) readline strace (4.22 -> 4.23) totem valgrind vlc xf86-video-savage xz (5.2.3 -> 5.2.4) === Details === ==== Mesa ==== Version update (18.1.1 -> 18.1.2) Subpackages: Mesa-dri-devel Mesa-libEGL1 Mesa-libGL1 Mesa-libglapi0 libgbm1 libwayland-egl1 - add mako_4_radv.patch - fixes Mako detection for RADV/intel - update to 18.1.2 * Fixes for libatomic checks on non-arm and non-x86 platforms * porting of additional libatomic checks to meson from autotools * numerous radv fixes * numerous intel fixes * A few fixes each for radeonsi, r300, ac, glx, and vulkan ==== Mesa-drivers ==== Version update (18.1.1 -> 18.1.2) Subpackages: Mesa-dri Mesa-dri-nouveau Mesa-gallium Mesa-libva libvdpau_nouveau libvdpau_r300 libvdpau_r600 libvdpau_radeonsi libvulkan_intel libvulkan_radeon libxatracker2 - add mako_4_radv.patch - fixes Mako detection for RADV/intel - update to 18.1.2 * Fixes for libatomic checks on non-arm and non-x86 platforms * porting of additional libatomic checks to meson from autotools * numerous radv fixes * numerous intel fixes * A few fixes each for radeonsi, r300, ac, glx, and vulkan ==== NetworkManager-applet ==== Version update (1.8.12 -> 1.8.14) Subpackages: NetworkManager-applet-lang NetworkManager-connection-editor libnma0 nma-data typelib-1_0-NMA-1_0 - Update to version 1.8.14: + Fix an applet crash on disconnect. + Fix VPN icons in AppIndicator mode. + Provide (limited) secret agent service even when a keyring is unavailable. + Fix saving of connections with "always ask" EAP secrets. + Register editor windows as GtkApplication windows, so that the editor quits at the right time with --edit option. + The AppStream metadata is now in /usr/share/metainfo as recommended by the specification. + Updated translations. - Drop nma-fix-applet-segfault.patch: Fixed upstream. ==== bison ==== Version update (3.0.4 -> 3.0.5) Subpackages: bison-lang - bison 3.0.5: * C++: Fix support of 'syntax_error' * Fix various C++ warnings * Fix issues with C++ parsers where out-of-bounds errors could happen when a rule with an empty ride-hand side raises a syntax error - drop gcc7-fix.patch, no longer required ==== discount ==== Version update (2.2.2 -> 2.2.3a) - Update to version 2.2.3a: * No changelog provided - Refresh patches: * discount-disable_ldconfig.patch * discount-fix-compile-warings.diff ==== dracut ==== - Add fix to override ACPI tables via initrd, a kernel config variable changed name (bsc#1098448) This is mainline git commit 940169e8d8e500498a3f350b2b3f341ae6548492 A 0580-check_for_CONFIG_ACPI_TABLE_UPGRADE.patch ==== ethtool ==== Version update (4.16 -> 4.17) - Update to new upstream release 4.17 * Fix: In ethtool.8, remove superfluous and incorrect \c * Fix: fix uninitialized return value * Fix: fix RING_VF assignment * Fix: remove unused global variable * Fix: several fixes in do_gregs() * Fix: correctly free hkey when get_stringset() fails * Fix: remove unreachable code * Fix: fix stack clash in do_get_phy_tunable and do_set_phy_tunable * Feature: Add register dump support for MICROCHIP LAN78xx ==== gcc8 ==== Version update (8.1.1+r260570 -> 8.1.1+r261583) Subpackages: cpp8 gcc8-c++ gcc8-fortran gcc8-info gcc8-locale gcc8-objc libasan5 libatomic1 libgcc_s1 libgcc_s1-32bit libgfortran5 libgomp1 libitm1 liblsan0 libmpx2 libmpxwrappers2 libobjc4 libquadmath0 libstdc++6 libstdc++6-32bit libstdc++6-devel-gcc8 libtsan0 libubsan1 - Update to gcc-8-branch head (r261583). * Includes fix for C++ ABI issue with passing a class with a trivial move constructor and deleted copy constructor. [GCC#86094] - riscv-builtin-eh-return.patch: Don't clobber retval when __builtin_eh_return called ==== git ==== Version update (2.17.1 -> 2.18.0) Subpackages: git-core git-cvs git-daemon git-email git-gui git-svn git-web gitk - git 2.18.0: * improvements to rename detection logic * When built with more recent cURL, GIT_SSL_VERSION can now specify "tlsv1.3" as its value. * "git mergetools" learned talking to guiffy. * various other workflow improvements and fixes * performance improvements and other developer visible fixes - drop git-bash-completion-egrep-color-fix.diff no longer required ==== gstreamer-rtsp-server ==== - Switch to meson build system: + Add meson, pkgconfig(glib-2.0),pkgconfig(gstreamer-app-1.0), pkgconfig(gstreamer-net-1.0), pkgconfig(gstreamer-rtp-1.0), pkgconfig(gstreamer-rtsp-1.0) and pkgconfig(gstreamer-sdp-1.0) BuildRequires. + Add meson macros, replacing autotools ones. + Pass disable_introspection=false, with-package-name='openSUSE GStreamer-rtsp-server package', with-package-origin='http://download.opensuse.org' and tests=false and examples=false to meson, ensure we build the features we want. Tests have always been disabled, be explicit about it, as they need a working network connection. + Drop pkgconfig(gstreamer-plugins-base-1.0) BuildRequires. + No longer rm la files, not needed when building with meson. ==== gstreamer-transcoder ==== Version update (1.12.2 -> 1.14.0) Subpackages: libgsttranscoder-1_0-0 typelib-1_0-GstTranscoder-1_0 - Update to version 1.14.0: + Misc bug fixing compared to 1.12.2. - Add gtk-doc BuildRequires and pass disable_doc=false and disable_introspection=false to meson, ensure we build the features we want. Package the now built api documentations in devel package. ==== hplip ==== Subpackages: hplip-hpijs hplip-sane - require python3-Pillow for hp-scan to work ==== installation-images-Kubic ==== Version update (14.375 -> 14.376) - merge gh#openSUSE/installation-images#252 - Tidy module.config comment notes - Add mailbox modules (boo#1040492) - 14.376 ==== kcm_tablet ==== Version update (3.0.0 -> 3.1.0) Subpackages: kcm_tablet-lang - Update to version 3.1.0 - Major changes: * LibWacom support. This should get our number of supported tablets out of the box on par with GNOME. Right now built-in database and user-configured database still take preference over libwacom, but this should change in the future releases. * Calibration improvements: calibration of Cintiq devices should be vastly improved, UI now has options to lock mapped area proportions and manually fine-tune area's coordinates. Also, the dialog should now open on the correct screen and look fine with dark themes too. * Support for devices that report different USB IDs for touchscreen and may be detected as two different devices (e.g. Cintiq Companion Hybrid). These devices need to be described via Wacom Tablet finder tool first, though. - Other changes: * Logging has been ported to QLoggingCategory (so now users have to enable it in kdebugsettings). * Status LED support improvements (Vangelis Tasoulas, Jason Gerecke). * Touchscreen rotation should now follow tablet rotation. - New devices in built-in database: * Wacom Intuos Pro L (PTH-860, 2017) (Melvin Vermeeren) * Lenovo Yoga 520 (Felipe Abrahao) * ThinkPad Yoga 460 definition (Albert Astals Cid) * Wacom Intuos Pro 2 M (Jeff H) - Requires Qt 5.7.0 or higher now to compile successfully ==== krb5 ==== Subpackages: krb5-32bit - BSC#1021402 move %{_libdir}/krb5/plugins/tls/k5tls.so to krb5 package so it is avaiable for krb5-client as well. ==== libtirpc ==== Subpackages: libtirpc-netconfig libtirpc3 libtirpc3-32bit - fix socket leak introduced by change-rpc-protocol-version-order patch (bsc#1087925) - add 0001-Fix-regression-introduced-by-change-rpc-version-orde.patch - Revert binddynport changes as they break backward compatibility [brc#1562169]. - add 002-revert-binddynport.patch ==== libzypp ==== Version update (17.3.1 -> 17.4.0) - Flags: make it std=c++14 ready - Replace 'redhat-release' with generic 'system-release' (#120) - Ignore /var, /tmp and /proc in zypper ps (bsc#1096617) - Show GPGME version in log - version 17.4.0 (2) ==== mailman ==== Version update (2.1.26 -> 2.1.27) - update to 2.1.27 * Existing protections against malicious listowners injecting evil scripts into listinfo pages have had a few more checks added. JVN#00846677/JPCERT#97432283/CVE-2018-0618 * A few more error messages have had their values HTML escaped. JVN#00846677/JPCERT#97432283/CVE-2018-0618 * The hash generated when SUBSCRIBE_FORM_SECRET is set could have been the same as one generated at the same time for a different list and IP address. While this is not thought to be exploitable in any way, the generation has been changed to avoid this. * An option has been added to bin/add_members to issue invitations instead of immediately adding members. * A new BLOCK_SPAMHAUS_LISTED_IP_SUBSCRIBE setting has been added to enable blocking web subscribes from IPv4 addresses listed in Spamhaus SBL, CSS or XBL. It will work with IPv6 addresses if Python's py2-ipaddress module is installed. The module can be installed via pip if not included in your Python. * Mailman has a new 'security' log and logs authentication failures to the various web CGI functions. The logged data include the remote IP and can be used to automate blocking of IPs with something like fail2ban. Since Mailman 2.1.14, these have returned an http 401 status and the information should be logged by the web server, but this new log makes that more convenient. Also, the 'mischief' log entries for 'hostile listname' noe include the remote IP if available. * admin notices of (un)subscribes now may give the source of the action. This consists of a %(whence)s replacement that has been added to the admin(un)subscribeack.txt templates. Thanks to Yasuhito FUTATSUKI for updating the non-English templates and help with internationalizing the reasons. * there is a new BLOCK_SPAMHAUS_LISTED_DBL_SUBSCRIBE setting to enable blocking web subscribes for addresses in domains listed in the Spamhaus DBL. * i18n & Bugfixes * for further details see NEWS ==== obs-service-source_validator ==== Version update (0.17 -> 0.18) - Update to version 0.18: * /usr/include/X11 is still a valid path. ==== perl-DateTime-Locale ==== Version update (1.200000 -> 1.220000) - updated to 1.22 see /usr/share/doc/packages/perl-DateTime-Locale/Changes 1.22 2018-06-08 - Update the glibc source data to the data from glibc 2.27. - Added documentation for this glibc data. - Reformatted the DateTime::Locale::FromData docs. 1.21 2018-06-08 - Added a new DateTime::Local->register_from_data method. This replaces the old custom locale registration system with something much simpler. Instead of providing a new class you simply provide a hash of data for the locale and it becomes available for use via DateTime::Local->load. Implemented by Ryley Breiddal. PR #21. Addresse GH #19. ==== readline ==== Subpackages: libreadline7 readline-devel readline-doc - Add patch readline-7.0-screen.patch to be able to parse settings in inputrc for all screen TERM variables starting with "screen." to fix boo#1095661 ==== strace ==== Version update (4.22 -> 4.23) - Update to strace 4.23 * Changes in behaviour * On x32 personality, 64-bit syscalls (such as readv) are now shown with "#64" suffix instead of "64:" prefix. * Changed error number output format: error numbers are consistently printed as "-1 ECONST (Error description)" for known errors and as "-1 (errno 123)" for unknown errors, regardless of -e raw qualification settings. * Unfetchable addresses inside arrays are now printed as comments. * Obsolete IA-32 mode syscall names are no longer printed on ia64. * Improvements * Implemented libdw backend for -k option, configured at build time using --with-libdw option. * Added -X option for configuring xlat output formatting (addresses Debian bug #692915). * Added support for personality designation ("64", "32", or "x32") to syscall qualifications in -e trace expressions. * Implemented injection of syscalls with no side effects as an alternative to injection of an invalid syscall (-e inject=SET:syscall= expression). * Improved support for reproducible builds (addresses Debian bug #896016). * Implemented decoding of BPF_PROG_QUERY and BPF_RAW_TRACEPOINT_OPEN bpf syscall commands. * Implemented decoding of INOTIFY_IOC_SETNEXTWD and PERF_EVENT_IOC_* ioctl commands. * Implemented decoding of PR_GET_SPECULATION_CTRL and PR_SET_SPECULATION_CTRL prctl syscall options. * Enhanced decoding of bpf, getsockopt, setsockopt, and socket syscalls. * Enhanced decoding of NETLINK_KOBJECT_UEVENT and NETLINK_ROUTE protocols. * Enhanced decoding of *_DIAG_MEMINFO netlink attributes. * Enhanced decoding of BTRFS_*, FS_IOC_*, SIOCGIFHWADDR, and SIOCSIFHWADDR ioctl commands. * Enhanced decoding of AF_BLUETOOTH socket addresses. * Implemented decoding of io_pgetevent syscall. * Wired up rseq syscall. * Updated lists of ADJ_*, BPF_*, BPF_F_*, BPF_PROG_TYPE_*, ETH_P_*, FPE_*, IFF_*, IPPROTO_*, MAP_*, MS_*, MSG_*, NETLINK_*, PACKET_*, PROT_*, SCTP_*, SECCOMP_FILTER_FLAG_*, SEGV_*, SEM_*, SHM_*, SW_*, V4L2_CID_*, V4L2_CTRL_CLASS_*, and V4L2_PIX_FMT_* constants. * Updated lists of ioctl commands from Linux 4.17. * Bug fixes * Fixed decoding of msgtyp argument of msgrcv syscall. * Fixed fallback definitions for F_*, MCL_*, MSG_ZEROCOPY, SI_*, and SOL_SOCKET constants. * Fixed -e trace=%pure expression on mips. - Use libdw for unwinding on Leap 15 or later ==== totem ==== Subpackages: nautilus-totem totem-lang totem-plugins - Align with what meson looks for: + Drop pkgconfig(gstreamer-plugins-bad-1.0) and pkgconfig(gstreamer-plugins-base-1.0) BuildRequires. + Add pkgconfig(gstreamer-pbutils-1.0) and pkgconfig(gstreamer-video-1.0) BuildRequires. ==== valgrind ==== - ad Implement-emulated-system-registers.-Fixes-392146.patch (bsc#1086543) ==== vlc ==== Subpackages: libvlc5 libvlccore9 vlc-codec-gstreamer vlc-lang vlc-noX vlc-qt - Require(post) vlc-noX for the -qt, -codecs, -jack and -vdpau sub-package: all those packages execute vlc-cache-gen (boo#1094893). ==== xf86-video-savage ==== - u_Add-check-for-max-HV-Value-to-ValidMode-hook.patch * Add check for max[HV]Value to ValidMode hook, since this functionality has been removed from xorg-server 1.20 ==== xz ==== Version update (5.2.3 -> 5.2.4) Subpackages: liblzma5 liblzma5-32bit xz-devel xz-lang - xz 5.2.4: * liblzma: - Allow 0 as memory usage limit instead of returning LZMA_PROG_ERROR. Now 0 is treated as if 1 byte was specified, which effectively is the same as 0. - Use "noexcept" keyword instead of "throw()" in the public headers when a C++11 (or newer standard) compiler is used. - Added a portability fix for recent Intel C Compilers. * xz: - Fix "xz --list --robot missing_or_bad_file.xz" which would try to print an unitialized string and thus produce garbage output. Since the exit status is non-zero, most uses of such a command won't try to interpret the garbage output. - "xz --list foo.xz" could print "Internal error (bug)" in a corner case where a specific memory usage limit had been set. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

1 0

[opensuse-factory] KMP build/packaging failure for i586
by Christof Hanke 28 Jun '18

28 Jun '18

Hi, I encounter a strange error building 3rd party KMP rpms. If you look at the x86_64 build : https://build.opensuse.org/package/live_build_log/openSUSE:Factory/openafs/… everything if fine. Using the same spec the 32bit i586 build fails in the actual packaging step: https://build.opensuse.org/package/live_build_log/openSUSE:Factory/openafs/… ... """ [ 2373s] Processing files: openafs-kmp-default-1.8.0_krpmbuild-1.19.i586 [ 2373s] error: File must begin with "/": * [ 2373s] error: File must begin with "/": Restart [ 2373s] error: File must begin with "/": config... [ 2373s] error: File must begin with "/": * [ 2373s] error: File must begin with "/": * [ 2373s] error: File must begin with "/": * [ 2373s] error: File must begin with "/": Linux/x86 [ 2373s] error: File must begin with "/": 4.17.1 [ 2373s] error: File must begin with "/": Kernel [ 2373s] error: File must begin with "/": Configuration [ 2373s] error: File must begin with "/": * [ 2373s] error: File must begin with "/": 64-bit [ 2373s] error: File must begin with "/": kernel [ 2373s] error: File must begin with "/": (64BIT) [ 2373s] error: File must begin with "/": [Y/n/?] [ 2373s] error: File must begin with "/": (NEW) [ 2373s] error: File must begin with "/": * """... Apparently, for the i586-case the automatically created file-list for the KMP is overwritten by the output of the kernel-configuration. Any ideas how to debug this ? Many thanks, Christof -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

6 22