September 2016 - openSUSE Factory - openSUSE Mailing Lists

[opensuse-factory] New Tumbleweed snapshot 20160912 released!
by Dominique Leuenberger 14 Sep '16

14 Sep '16

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: NetworkManager-gnome aqbanking (5.6.10 -> 5.7.1beta) ccache (3.2.7 -> 3.3.1) cronie curl (7.50.1 -> 7.50.2) ffmpeg flac gcin geoclue2 (2.4.3 -> 2.4.4) getdata (0.9.3 -> 0.9.4) gwenhywfar (4.15.3 -> 4.15.5beta) krita (3.0 -> 3.0.1) libvirt (2.1.0 -> 2.2.0) libvpx libzypp (16.2.2 -> 16.2.3) nghttp2 (1.13.0 -> 1.14.0) obs-service-tar_scm (0.6.1.1471594222.d257927 -> 0.6.1.1472657181.ff9a5ca) pcsc-lite (1.8.17 -> 1.8.18) python3-cairo (1.10.0 -> 1.10.1~git20121222.5a45b0b) sg3_utils (1.42 -> 1.43) suse-module-tools sysconfig (0.83.9 -> 0.84.0) update-alternatives (1.18.4 -> 1.18.10) util-linux (2.28.1 -> 2.28.2) util-linux-systemd (2.28.1 -> 2.28.2) wireshark (2.0.5 -> 2.2.0) yast2-auth-client (3.3.10 -> 3.3.11) yast2-drbd (3.1.24 -> 3.1.26) zypper (1.13.9 -> 1.13.10) === Details === ==== NetworkManager-gnome ==== Subpackages: NetworkManager-connection-editor libnm-gtk0 typelib-1_0-NMGtk-1_0 - Drop nm-applet-fix-missing-translation.patch: This patch leads to conflict between accelerate keys. And it won't fix the missing translations, since the problem comes from those untranslated POs. ==== aqbanking ==== Version update (5.6.10 -> 5.7.1beta) Subpackages: aqbanking-devel aqbanking-lang - quick fix for $CPP being unset and configure failing to handle include dirs properly - Update to 5.7.1beta: + re-retrieval of the account list is not required anymore to continue using existing accounts - Update to 5.7.0beta: + finally fixes the problem of not correlating accounts correctly if the bank returns multiple accounts with the same account number and IBAN - Update to 5.6.12 + Improvement in handling account owners ==== ccache ==== Version update (3.2.7 -> 3.3.1) - Update to version 3.3.1: * Fixed a problem in the ?multiple -arch options? support introduced in 3.3. When using the direct mode (the default), different combinations of -arch options were not detected properly. * Fixed an issue when compiler option -Wp,-MT,path is used instead of -MT path (and similar for -MF, -MP and -MQ) and run_second_cpp (CCACHE_CPP2) is enabled. - Changes for version 3.3.0: * A C99-compatible compiler is now required to build ccache. * The configuration option run_second_cpp (CCACHE_CPP2) now defaults to true. This improves ccache?s out-of-the-box experience for compilers that can?t compile their own preprocessed output with the same outcome as if they compiled the real source code directly, e.g. newer versions of GCC and Clang. * The configuration option hash_dir (CCACHE_HASHDIR) now defaults to true. * Added a new ignore_headers_in_manifest configuration option, which specifies headers that should be ignored in the direct mode. * Added a new prefix_command_cpp (CCACHE_PREFIX_CPP) configuration option, which specifies one or several prefixes to add to the command line ccache uses when invoking the preprocessor. * Added a new limit_multiple (CCACHE_LIMIT_MULTIPLE) configuration option, which specifies how much of the cache to remove when cleaning. * Added a new keep_comments_cpp (CCACHE_COMMENTS) configuration option, which tells ccache not to discard the comments before hashing preprocessor output. This can be used to check documentation with -Wdocumentation. * Added a new sloppiness option no_system_headers, which tells ccache not to include system headers in manifest files. * Added a new statistics counter that tracks the number of performed cleanups due to the cache size being over the limit. The value is shown in the output of ?ccache -s?. * Added support for relocating debug info directory using - fdebug-prefix-map. This allows for cache hits even when hash_dir is used in combination with base_dir. * Added a new ?cache hit rate? field to the output of ?ccache -s?. * Added support for caching compilation of assembler code produced by e.g. ?gcc -S file.c?. * Added support for cuda including the -optf/--options-file option. * Added support for Fortran 77. * Added support for multiple -arch options to produce "fat binaries". * Multiple identical -arch arguments are now handled without bailing. * The concatenated form of some long compiler options is now recognized, for example when using -isystemPATH instead of - isystem PATH. * If hard-linking is enabled and but fails (e.g. due to cross-device linking), ccache now falls back to copying instead of running the compiler. * Made the hash_dir option only have effect when generating debug info. * ccache now knows how to convert absolute paths to relative paths inside dependency files when using base_dir. * Improved parsing of -g* options. * Made ccache understand -Wp,-D* options. * ccache now understands the undocumented -coverage (only one dash) GCC option. * Names of included files are no longer included in the hash of the compiler?s preprocessed output. This leads to more potential cache hits when not using the direct mode. * Increased buffer size used when reading file data. This improves performance slightly. - Changes for version 3.2.8: * Bail out on too hard compiler option -P. * Fixed clang test suite when running on Linux. * Fixed build and test for MinGW32 and Windows. - Add fortran amd objc symlinks ==== cronie ==== Subpackages: cron - Add fix for bnc#983925 to run crons even when not on AC_POWER * Nowdays it does not make much sense to not run crons when on battery and actually it can even confuse people ==== curl ==== Version update (7.50.1 -> 7.50.2) Subpackages: libcurl-devel libcurl4 - update to 7.50.2 Bugfixes: * mbedtls: Added support for NTLM * SSH: fixed SFTP/SCP transfer problems * multi: make Curl_expire() work with 0 ms timeouts * mk-ca-bundle.pl: -m keeps ca cert meta data in output * TFTP: Fix upload problem with piped input * CURLOPT_TCP_NODELAY: now enabled by default * mbedtls: set verbose TLS debug when MBEDTLS_DEBUG is defined * http2: always wait for readable socket * cmake: Enable win32 large file support by default * cmake: Enable win32 threaded resolver by default * winbuild: Avoid setting redundant CFLAGS to compile commands * curl.h: make CURL_NO_OLDIES define CURL_STRICTER * docs: make more markdown files use .md extension * docs: CONTRIBUTE and LICENSE-MIXING were converted to markdown * winbuild: Allow changing C compiler via environment variable CC * rtsp: accept any RTSP session id * HTTP: retry failed HEAD requests on reused connections too * configure: add zlib search with pkg-config * openssl: accept subjectAltName iPAddress if no dNSName match * MANUAL: Remove invalid link to LDAP documentation * socks: improved connection procedure * proxy: reject attempts to use unsupported proxy schemes * proxy: bring back use of "Proxy-Connection:" * curl: allow "pkcs11:" prefix for client certificates * spnego_sspi: fix memory leak in case *outlen is zero * SOCKS: improve verbose output of SOCKS5 connection sequence * SOCKS: display the hostname returned by the SOCKS5 proxy server * http/sasl: Query authentication mechanism supported by SSPI before using * sasl: Don't use GSSAPI authentication when domain name not specified * win: Basic support for Universal Windows Platform apps * nss: fix incorrect use of a previously loaded certificate from file, https://curl.haxx.se/docs/adv_20160907.html * nss: work around race condition in PK11_FindSlotByName() * ftp: fix wrong poll on the secondary socket * openssl: build warning-free with 1.1.0 (again) * HTTP: stop parsing headers when switching to unknown protocols * test219: Add http as a required feature * TLS: random file/egd doesn't have to match for conn reuse * schannel: Disable ALPN for Wine since it is causing problems * http2: make sure stream errors don't needlessly close the connection * http2: return CURLE_HTTP2_STREAM for unexpected stream close * darwinssl: --cainfo is intended for backward compatibility only * speed caps: not based on average speeds anymore * configure: make the cpp -P detection not clobber CPPFLAGS * http2: use named define instead of magic constant in read callback * http2: skip the content-length parsing, detect unknown size * http2: return EOF when done uploading without known size * darwinssl: test for errSecSuccess in PKCS12 import rather than noErr * openssl: fix CURLINFO_SSL_VERIFYRESULT ==== ffmpeg ==== Subpackages: libavcodec57 libavformat57 libavutil55 libswresample2 libswscale4 - Fix linking errors for arm and aarch64 builds, boo#995937 * Add ffmpeg-fix-CONFIG_VC1DSP-changes.patch ==== flac ==== Subpackages: flac-devel libFLAC++6 libFLAC8 libFLAC8-32bit - Drop patch that should be upstreamed first, otherwise we will have to keep it ofrever: * flac-ocloexec.patch - Drop wrong patch: * flac-fix-pkgconfig.patch + If using this change you get assert.h include overriden in your project by the one from FLAC/ which is not what upstream desired If packages fail to build they should fix their include ==== gcin ==== Subpackages: gcin-gtk2 gcin-gtk3 gcin-qt4 gcin-qt5 libgcin-im-client1 - Remove libappindicator since KDE already brought system tray icon back. ==== geoclue2 ==== Version update (2.4.3 -> 2.4.4) Subpackages: typelib-1_0-Geoclue-2_0 - Update to version 2.4.4: + Expose location timestamp over D-Bus. + Add a TimeThreshold property that allows clients to limit location updates by time. + Annotate default value of DistanceThreshold property. + Check for full internet availability before declaring ourselves online. + Fix autogen.sh against builddir ? srcdir. + Fix a bunch of memory leaks. + Fix a bunch of build issues. + Other minor fixes. ==== getdata ==== Version update (0.9.3 -> 0.9.4) Subpackages: getdata-devel getdata-doc libgetdata++6 libgetdata7 - update to 0.9.4 for all relevant changes see https://sourceforge.net/projects/getdata/files/getdata/0.9.4/ * BUG FIX: Arbitrarily-long reads of FLAC-encoded files now work. Previously, each FLAC frame was written to the start of the output buffer, overwriting the previous frame, and leaving most of the buffer uninitialised. Reported by S. J. Benton. * BUG FIX: Data read from FLAC-encoded 1-byte types are now correct. Previously, although all data requested was read, only the first half would be returned, encoded as 16-bit data (i.e., with a zero-byte between every sample). ==== gwenhywfar ==== Version update (4.15.3 -> 4.15.5beta) Subpackages: gwenhywfar-devel gwenhywfar-lang libgwengui-cpp0 libgwengui-fox16-0 libgwengui-gtk2-0 libgwengui-qt4-0 libgwenhywfar60 libgwenhywfar60-plugins - quick fix for $CPP being unset and configure failing to handle include dirs properly - Update to 4.15.5beta + Fix connections to buggy SSL servers ==== krita ==== Version update (3.0 -> 3.0.1) - Do not install the BreezeDark and BreezeHighContrast color schemes, they are already included in breeze5-style (kde#368093) - Update to version 3.0.1 For a list of changes see: https://krita.org/en/item/krita-3-0-1-update-brings-numerous-fixes/ - Add lang subpackage ==== libvirt ==== Version update (2.1.0 -> 2.2.0) Subpackages: libvirt-client libvirt-daemon libvirt-daemon-config-network libvirt-daemon-config-nwfilter libvirt-daemon-driver-interface libvirt-daemon-driver-libxl libvirt-daemon-driver-lxc libvirt-daemon-driver-network libvirt-daemon-driver-nodedev libvirt-daemon-driver-nwfilter libvirt-daemon-driver-qemu libvirt-daemon-driver-secret libvirt-daemon-driver-storage libvirt-daemon-driver-uml libvirt-daemon-driver-vbox libvirt-daemon-lxc libvirt-daemon-qemu libvirt-daemon-xen libvirt-libs - Update to libvirt 2.2.0 - Many incremental improvements and bug fixes, see http://libvirt.org/news.html - Dropped patches: 856965b3-qemu-secdriver.patch, 541e9ae6-cpu-vendor-crash-fix.patch, d53d4650-qemu-rbd-auth.patch ==== libvpx ==== Subpackages: libvpx4 libvpx4-32bit - Ensure the enable vp8 vp9 postproc and multithread are always on - Enable verbose build - Add switches to be able to build chromium: * --enable-experimental --enable-spatial-svc ==== libzypp ==== Version update (16.2.2 -> 16.2.3) - xmlout::Node: Allow adding nore attributes while the start node is not closed (FATE#320699) - version 16.2.3 (0) ==== nghttp2 ==== Version update (1.13.0 -> 1.14.0) - Update to version 1.14.0: * lib: Make emit_header() return void since it always succeed * lib: Add nghttp2_hd_deflate_hd_vec() deflate API to support multiple buffer input * lib: since hd_inflate_commit_indexed() always return 0, remove the return value check in nghttp2_hd_inflate_hd_nv() * lib: Use memeq() instead of lstreq() in lookup_token() * lib: More strict stream state handling * lib: Modify genlibtokenlookup.py to remove redundant header comparisons and remove inline qualifier of lookup_token() in genlibtokenlookup.py * lib: Fix wrong tree operation to avoid cycle * lib: Make get_max_index() return the max index in frame, so we don't need to do extra calculation * lib: Add nghttp2_on_invalid_header_callback * lib: Log frame's stream ID for header debug logging * doc: Remove old doc about differential encoding in HPACK * doc: Document about ALPN in nghttpx howto * nghttpx: Log error code from getsockopt(SO_ERROR) on first write event * nghttpx: Don't change pushed stream's priority * nghttpx: Log backend connection failure in WARN level * nghttpx: Fix bug that api and healthmon parameters do not work with http2 proxy * nghttpx: Add access log variable for backend host and port * nghttpx: Use copy instead of const reference of backend group * nghttpx: Reload configuration with SIGHUP * nghttp: Adjust weight according to Firefox stable * nghttp: Call error callback when invalid header field is received and ignored * nghttp: Allow multiple -p option * deflatehd: Call nghttp2_hd_deflate_change_table_size only if table size is changed from default ==== obs-service-tar_scm ==== Version update (0.6.1.1471594222.d257927 -> 0.6.1.1472657181.ff9a5ca) - Update to version 0.6.1.1472657181.ff9a5ca: * make yaml dependency optional for travis * Add an option 'master' to submodules to fetch the latest mater branch. * tar_scm.py: Always use an absolute path for the 'output' directory * do not set files back to 1970... * transfer mtime via obsinfo * create needed sub directories for cache handling * prefer local cache directory * - weak dependency to mercurial for debian * add python-dateutil require for debian * testing on debian is broken atm - Update to version 0.6.1.1472656157.58c52d8: * make yaml dependency optional for travis * Add an option 'master' to submodules to fetch the latest mater branch. * tar_scm.py: Always use an absolute path for the 'output' directory * do not set files back to 1970... * transfer mtime via obsinfo * create needed sub directories for cache handling * prefer local cache directory * - weak dependency to mercurial for debian * add python-dateutil require for debian * testing on debian is broken atm - Update to version 0.6.1.1472655284.4930b81: * add python-dateutil require for debian * testing on debian is broken atm - fix debian builds - Update to version 0.6.1.1472636708.355b59a: * fix some more merge regressions * pep8 fixes/workarounds * make yaml dependency optional for travis * Add an option 'master' to submodules to fetch the latest mater branch. * tar_scm.py: Always use an absolute path for the 'output' directory * do not set files back to 1970... * transfer mtime via obsinfo * create needed sub directories for cache handling * prefer local cache directory * - weak dependency to mercurial for debian ==== pcsc-lite ==== Version update (1.8.17 -> 1.8.18) Subpackages: libpcsclite1 - Version update to 1.8.18: * SCardDisconnect(): much faster with SCARD_UNPOWER_CARD * SCardConnect(): Fix a possible duplicated hCard context * Fix compilation on FreeBSD * Fix compilation on Solaris * Some other minor improvements - Obsolete the conditionals for sle11 as it does not build there and ends unresolvable anyway - Remove the presets magic after discussion with standa as new macros actually work around this correctly ==== python3-cairo ==== Version update (1.10.0 -> 1.10.1~git20121222.5a45b0b) - pycairo didn't see new release for 4 years, so git snapshot is reasonable. - update to version 1.10.1~git20121222.5a45b0b: * Add support for RecordingSurface * Create a setup.py * Update tests * Implement ImageSurface.get_data() * Implement ImageSurface.create_for_data() * Remove option to use xpyb since it does not currently support Python 3, and the .pc file name conflicts with the Python 2 version. * Add some new constants, methods and functions * Add support for cairo_region_t * Update waf to 1.7.8 * Correct the pycairo version number in config.h - drop waf-py34.patch, upstreamed - Update to GNOME 3.20.2 FATE#318572 ==== sg3_utils ==== Version update (1.42 -> 1.43) - using "--lun" now works correctly (bsc#981452), adding patch: * add-lunsearch-filter-to-searchexisting.patch - Update to new upstream release 1.43: * sg_senddiag: add --timeout=SEC option * sg_sanitize: add --timeout=SEC option * sg_format: add --timeout=SEC option * sg_decode sense: add --cdb option * sg_ses: handle 2 bit EIIOE field in aes dpage - expand join handling of SAS connectors and others - expand join debug code * sg_logs: fix volume statistics lpage when subpage is zero (ssc5r02a); decode mount history log parameter - decode Requested recovery, TapeAlert response, and Service buffer information lpages for tape - add '--pdt=DT' option * sg_inq: fix potential unbounded loop in --export - update version descriptor list to 20160420 * sg_inq+sg_vpd: update Extended inquiry data vpd page (spc5r09) - add --force option to bypass checking supported vpd pages page and fetch requested page directly * sg_reassign+sg_write_same: fix ULONG_MAX problem * sg_turs+sg_requests: make both accept '--num=NUM' and '--number=NUM' for mutual compatibility * sg_zone: fix debug cdb naming * sg_opcode: add '--enumerate' and '--pdt=' options * sg_raw: add '--enumerate' option * sg_lib: add SSC maintenance in/out sa names - add read buffer(16) command mode names - add sg_decode_transportid_str() * sg_lib_data: sync asc/ascq codes with T10 20160425 * rescan-scsi-bus.sh: harden code * clang --analyze static checker clean ups * shellcheck cleanup on scripts * automake: add AM_PROG_AR to configure.ac - upgrade to version 1.15 - Include patches from upstream: * sg_inq+sg_vpd: add --force option to bypass checking of supported vpd pages * sg_lib_data: add Bind and Unbind to service action out(16) - Remove 0001-sg_vpd-sg_inq-Safe-VPD-page-access.patch; merged with upstream. - sg_inq,sg_vpd: Safe VPD page access (bsc#945094) * Add 0001-sg_vpd-sg_inq-Safe-VPD-page-access.patch ==== suse-module-tools ==== - blacklist efifb on aarch64 (bsc#996080) ==== sysconfig ==== Version update (0.83.9 -> 0.84.0) Subpackages: sysconfig-netconfig - version 0.84.0 - netconfig: intoduced a batch mode (bsc#989741) It permits to preform multiple modify/remove operations and call the effective update at the end of the batch. - netconfig: use ntpq, as ntpdc is deprecated (bsc#894889), lower sleep lock wait granularity and fix NTP_KEYS parsing - Drop the omc svcinfo data as they are not used anymore - Run spec-cleaner to remove few obsolete things - Drop cvs helper files as we are in git - Rename configure.in to configure.ac ==== update-alternatives ==== Version update (1.18.4 -> 1.18.10) - Version update to 1.18.10: * Various few fixes, see changelog file in the package * Refresh patch update-alternatives-suse.patch * Remove patch dpkg-sparc.diff - Use ls . instead of * to avoid error when nothing is in alternatives directory ==== util-linux ==== Version update (2.28.1 -> 2.28.2) Subpackages: libblkid-devel libblkid1 libblkid1-32bit libfdisk1 libmount1 libmount1-32bit libsmartcols1 libuuid-devel libuuid1 libuuid1-32bit - Update to version 2.28.2, bugfix release, see ftp://ftp.kernel.org/pub/linux/utils/util-linux/v2.28/v2.28.2-ReleaseNotes ==== util-linux-systemd ==== Version update (2.28.1 -> 2.28.2) - Update to version 2.28.2, bugfix release, see ftp://ftp.kernel.org/pub/linux/utils/util-linux/v2.28/v2.28.2-ReleaseNotes ==== wireshark ==== Version update (2.0.5 -> 2.2.0) - Wireshark 2.2.0: * Bug fixes and updated protocol support as listed in: https://www.wireshark.org/docs/relnotes/wireshark-2.2.0.html * Drop wireshark-pkg-config.patch because code is now in upstream * Drop wireshark-1.12.6-fix-QT-PIC-PIE.patch because no more need for build. And create errors at build Tumbleweed. * Rebase wireshark-1.10.0-authors-pod2man.patch - Wireshark 2.0.6: This release fixes a number of issues that made it possible to make Wireshark crash by injecting a malformed packet onto the wire or by convincing someone to read a malformed packet trace file. (bsc#998099) * The QNX6 QNET dissector could crash wnpa-sec-2016-50 CVE-2016-7175 * The H.225 dissector could crash wnpa-sec-2016-51 CVE-2016-7176 * The Catapult DCT2000 dissector could crash wnpa-sec-2016-52 CVE-2016-7177 * The UMTS FP dissector could crash wnpa-sec-2016-53 CVE-2016-7178 * The Catapult DCT2000 dissector could crash wnpa-sec-2016-54 CVE-2016-7179 * The IPMI Trace dissector could crash wnpa-sec-2016-55 CVE-2016-7180 * Further bug fixes and updated protocol support as listed in: https://www.wireshark.org/docs/relnotes/wireshark-2.0.6.html - fix permissions warning on appdata, does not need +x ==== yast2-auth-client ==== Version update (3.3.10 -> 3.3.11) - Fix an issue with UI layout after leaving a domain (bsc#997380). - Make DNS update optional for AD enrollment (bsc#991564). - Fix inaccurate description of ad_server parameter (bsc#996873). - Do not allow using hard bind policy in LDAP config (bsc#996538). Bump version to 3.3.11. ==== yast2-drbd ==== Version update (3.1.24 -> 3.1.26) - bsc#997838, fix error of rename a exist drbd resource. - Version 3.1.26 - fate#321158, remove automatic change lvm filter - Version 3.1.25 ==== zypper ==== Version update (1.13.9 -> 1.13.10) Subpackages: zypper-aptitude zypper-log - products --xmlfwd: Literally forward the XML tags found in a product file (FATE#320699) - BuildRequires: libzypp-devel >= 16.2.3 - man: enhance install/uninstall section - version 1.13.10 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

1 0

Re: [opensuse-factory] looking for script hacker to update online release notes script
by Ludwig Nussel 14 Sep '16

14 Sep '16

Carl Xiong schrieb: > On Fri, 09 Sep 2016 21:05:22 +0800, Ludwig Nussel <ludwig.nussel(a)suse.de> wrote: >> The release notes pages on doc.opensuse.org are created by an old >> shell script¹ by downloading rpms and extracting them in a >> directory. >> After switching from fetching only released packages to ones >> directly created from github² the script now confuses packages from >> different distributions. So e.g. the 42.1 page actually displays >> 42.2³. > > Hi, I'd like to help improve the script. > > However, after I took a close look at the script, I reached the > conclusion that the script has no serious issue. The "confusion" for > 'openSUSE_Leap_42.1' is not the script's fault but the corrupted state > of the build repo at 'Documentation:Auto/openSUSE_Leap_42.1'. > > See [1], there > 'release-notes-openSUSE-42.2.20160728.e297df1-4.1.noarch.rpm' is still > listed. The script downloads package through API link like > 'https://api.opensuse.org/build/Documentation:Auto/openSUSE_Leap_42.1/x86_64…' > [2], which seems to point to the package with the _largest_ release > number. In Leap42.1 case, it sees '42.2.*' larger than any '41.1.*'. > So I believe refreshing/rebuilding the build repo would resolve the > problem. The 42.2.* package might be some residual from transitions. You are right. Looks like the 42.2 release notes were accidentally built for 42.1 once and then got build disabled. The old binaries were still floating around though. Wipebinaries helped. Thanks! cu Ludwig -- (o_ Ludwig Nussel //\ V_/_ http://www.suse.com/ SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) -- 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] openQA first install and test remains scheduled
by Normand 14 Sep '16

14 Sep '16

Hi there, I installed for the first time openQA in a PowerPC Tumbleweed machine as per doc in (1) (built from devel:openQA ppc64le TW) Then I copied an iso and started a test as per doc in (2) no error was reported. BUT the WEBUI list the tests as "scheduled" Is there a log I could check to understand why the tests are not sent to worker in idle state (3) and remain scheduled ? (1) https://github.com/os-autoinst/openQA/blob/master/docs/Installing.asciidoc https://github.com/os-autoinst/openQA/blob/master/docs/GettingStarted.ascii… (2) https://github.com/os-autoinst/openQA/blob/master/docs/GettingStarted.ascii… === # ls -l /var/lib/openqa/share/factory/iso/ -rw-r--r-- 1 root root 4176916480 Sep 13 11:47 openSUSE-Tumbleweed-DVD-ppc64le-Snapshot20160907-Media.iso # /var/lib/openqa/script/client isos post ISO=openSUSE-Tumbleweed-DVD-ppc64le-Snapshot20160907-Media.iso DISTRI=opensuse VERSION=Tumbleweed FLAVOR=DVD ARCH=ppc64le BUILD=20160907 === (3) === abanc:~ # systemctl status openqa-worker@1 openqa-worker(a)1.service - openQA Worker #1 Loaded: loaded (/usr/lib/systemd/system/openqa-worker@.service; disabled; vendor preset: disabled) Active: active (running) since Mon 2016-09-12 17:29:00 CEST; 20h ago Process: 22642 ExecStartPre=/usr/bin/install -d -m 0755 -o _openqa-worker /var/lib/openqa/pool/%i (code=exited, status=0/SUCCESS) Main PID: 22645 (worker) Tasks: 1 (limit: 512) CGroup: /system.slice/system-openqa\x2dworker.slice/openqa-worker(a)1.service └─22645 /usr/bin/perl /usr/share/openqa/script/worker --instance 1 === -- Michel Normand -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

4 5

Re: [opensuse-factory] openQA first install and test remains scheduled
by Normand 14 Sep '16

14 Sep '16

On 14/09/2016 15:40, Ondrej Holecek wrote: > On středa 14. září 2016 15:21:41 CEST Normand wrote: >> ... >> >> I confirmed that submitted tests remain scheduled as long as the workers are >> not manually restarted. with following scenario: >> submit 5 jobs ids 47 .. 51 => all 5 remain idles >> restart manually two workers => 5 jobs are handled (and failed) >> submit 5 jobs ids 52 .. 56 => all 5 remain idles >> capture the last failing autoinst-log.txt (of id 51) >> and the two worker status in attached log file > Are you sure you have update workers? What is the status of your websockets > server (openqa-websockets service). > Workers wait for notification from openqa about new available jobs. not sure I understand the question about "update workers" ? do you refer to the rpm version ? it is last one from devel:openQA built for TW ppc64le (1) the openqa-websockets is reported as active but no specific log (2) > >> >> The systemctl status output is reporting an unknown command error. >> I assume this is the cause of the tests remaining scheduled. >> >> Note that in addition to this problem there are two other problems: >> 2- Can't open directory /proc/device-tree/cpus/ > This looks like QEMU problem, is qemu working for you fine when started > manually? I previously verified that qemu working while using it to build a package (osc build --vm-type kvm) I will try to execute qemu with same parameters as extracted from previous log to check if that make a difference. > >> 3- failed to find romfile "vgabios-stdvga.bin" > > There is also problem with isotovideo's command thread exiting just after > start. Given the uninitialized variables and their positions I think you are > running some old version. > as reported in (1) the isotovideo is from last os-autoinst version (1) https://build.opensuse.org/package/show/home:michel_mno:branches:devel:open… === $rpm -qa |grep -iE 'autoinst|openqa' openQA-common-4.4.1473439595.7ea7b0d-198.1.noarch os-autoinst-4.4.1473579745.e87b5e4-97.1.ppc64le openQA-client-4.4.1473439595.7ea7b0d-198.1.noarch openQA-4.4.1473439595.7ea7b0d-198.1.noarch openQA-worker-4.4.1473439595.7ea7b0d-198.1.noarch === $rpm -qf /usr/bin/isotovideo os-autoinst-4.4.1473579745.e87b5e4-97.1.ppc64le === (2) after host reboot, workers manually restarted, and 5 submitted jobs: === $sudo systemctl status -n1000 openqa-worker@2 * openqa-worker(a)2.service - openQA Worker #2 Loaded: loaded (/usr/lib/systemd/system/openqa-worker@.service; disabled; vendor preset: disabled) Active: active (running) since Wed 2016-09-14 15:58:12 CEST; 11min ago Process: 2593 ExecStartPre=/usr/bin/install -d -m 0755 -o _openqa-worker /var/lib/openqa/pool/%i (code=exited, status=0/SUCCESS) Main PID: 2595 (worker) Tasks: 1 (limit: 512) CGroup: /system.slice/system-openqa\x2dworker.slice/openqa-worker(a)2.service `-2595 /usr/bin/perl /usr/share/openqa/script/worker --instance 2 Sep 14 15:58:12 abanc systemd[1]: Starting openQA Worker #2... Sep 14 15:58:12 abanc systemd[1]: Started openQA Worker #2. [michel@abanc:~] $sudo systemctl status -n1000 openqa-worker@1 * openqa-worker(a)1.service - openQA Worker #1 Loaded: loaded (/usr/lib/systemd/system/openqa-worker@.service; disabled; vendor preset: disabled) Active: active (running) since Wed 2016-09-14 15:58:09 CEST; 11min ago Process: 2578 ExecStartPre=/usr/bin/install -d -m 0755 -o _openqa-worker /var/lib/openqa/pool/%i (code=exited, status=0/SUCCESS) Main PID: 2582 (worker) Tasks: 1 (limit: 512) CGroup: /system.slice/system-openqa\x2dworker.slice/openqa-worker(a)1.service `-2582 /usr/bin/perl /usr/share/openqa/script/worker --instance 1 Sep 14 15:58:09 abanc systemd[1]: Starting openQA Worker #1... Sep 14 15:58:09 abanc systemd[1]: Started openQA Worker #1. [michel@abanc:~] $systemctl status -n100 openqa-websockets * openqa-websockets.service - The openQA WebSockets server Loaded: loaded (/usr/lib/systemd/system/openqa-websockets.service; enabled; vendor preset: disabled) Active: active (running) since Wed 2016-09-14 15:51:31 CEST; 18min ago Main PID: 1896 (openqa-websocke) Tasks: 1 (limit: 512) CGroup: /system.slice/openqa-websockets.service `-1896 /usr/bin/perl /usr/share/openqa/script/openqa-websockets Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable. === -- Michel Normand -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

1 0

Re: [opensuse-factory] openQA first install and test remains scheduled
by Normand 13 Sep '16

13 Sep '16

On 13/09/2016 14:51, Ondrej Holecek wrote: > On úterý 13. září 2016 14:27:55 CEST Normand wrote: >> Hi there, >> I installed for the first time openQA in a PowerPC Tumbleweed machine >> as per doc in (1) (built from devel:openQA ppc64le TW) >> Then I copied an iso and started a test as per doc in (2) no error was >> reported. BUT the WEBUI list the tests as "scheduled" >> >> Is there a log I could check to understand why the tests are not sent to >> worker in idle state (3) and remain scheduled ? > There is no log for this, but I suspect mismatch in worker classes. Check if > your worker WORKER_CLASS (you can find it in Admin->Workers menu) match > WORKER_CLASS variable in tests (Test overview -> Settings or Test overview -> > Logs&Assets -> vars.json). > > Worker's WORKER_CLASS is usually computed automatically. Tests WORKER_CLASS is > by default empty, which matches everything. > > Usual place to put WORKER_CLASS for tests is in Machines table (Admin -> > Machines) which you then use in you job group. Thank you Ondrej, but it seems to be same WORKER_CLASS: as per extracted data the WORKER_CLASS qemu_ppc64le of the test is supported by the worker that has: qemu_ppc64le,qemu_ppc64,qemu_ppc === worker properties: CPU_ARCH ppc64le CPU_MODELNAME POWER8 (raw), altivec supported INTERACTIVE 0 INTERACTIVE_REQUESTED 0 MEM_MAX 261533 STOP_WAITFORNEEDLE 0 STOP_WAITFORNEEDLE_REQUESTED 0 WORKER_CLASS qemu_ppc64le,qemu_ppc64,qemu_ppc === tests/14#settings ARCH ppc64le BACKEND qemu BUILD 20160907 DESKTOP textmode DISTRI opensuse DVD 1 FLAVOR DVD ISO openSUSE-Tumbleweed-DVD-ppc64le-Snapshot20160907-Media.iso ISO_MAXSIZE 4700372992 MACHINE ppc64le NAME 00000014-opensuse-Tumbleweed-DVD-ppc64le-Build20160907-textmode@ppc64le NOIMAGES 1 OFW 1 QEMUCPU host TEST textmode VERSION Tumbleweed VIDEOMODE text WORKER_CLASS qemu_ppc64le === >> >> >> (1) >> https://github.com/os-autoinst/openQA/blob/master/docs/Installing.asciidoc >> https://github.com/os-autoinst/openQA/blob/master/docs/GettingStarted.ascii >> doc#getting-tests (2) >> https://github.com/os-autoinst/openQA/blob/master/docs/GettingStarted.ascii >> doc#adding-a-new-iso-to-test === >> # ls -l /var/lib/openqa/share/factory/iso/ >> -rw-r--r-- 1 root root 4176916480 Sep 13 11:47 >> openSUSE-Tumbleweed-DVD-ppc64le-Snapshot20160907-Media.iso # >> /var/lib/openqa/script/client isos post >> ISO=openSUSE-Tumbleweed-DVD-ppc64le-Snapshot20160907-Media.iso >> DISTRI=opensuse VERSION=Tumbleweed FLAVOR=DVD ARCH=ppc64le BUILD=20160907 >> === >> (3) >> === >> abanc:~ # systemctl status openqa-worker@1 >> openqa-worker(a)1.service - openQA Worker #1 >> Loaded: loaded (/usr/lib/systemd/system/openqa-worker@.service; >> disabled; vendor preset: disabled) Active: active (running) since Mon >> 2016-09-12 17:29:00 CEST; 20h ago Process: 22642 >> ExecStartPre=/usr/bin/install -d -m 0755 -o _openqa-worker >> /var/lib/openqa/pool/%i (code=exited, status=0/SUCCESS) Main PID: 22645 >> (worker) >> Tasks: 1 (limit: 512) >> CGroup: >> /system.slice/system-openqa\x2dworker.slice/openqa-worker(a)1.service └─22645 >> /usr/bin/perl /usr/share/openqa/script/worker --instance 1 === > -- Michel Normand -- 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] Reminder: openSUSE Leap 42.2 Beta 2 submission deadline Thursday
by Ludwig Nussel 13 Sep '16

13 Sep '16

Hi, This is a reminder for the upcoming openSUSE Leap 42.2 Beta 2 submission deadline. Please submit your packages latest Thursday, September 15 to openSUSE:Leap:42.2. If you've submitted already weeks ago but your submission wasn't accepted yet, please check why. The most common reasons are - Build failure in Staging:adi. All package submissions are built isolated in openSUSE:Leap:42.2:Staging:adi:<number> to make sure the submission is complete. Sometimes that fails due to different configurations in the devel project for example. Sometimes other packages need to be grouped into the same project. Let us know if your package needs manual help there. - Factory review. The 42.2 submissions wait for the submission to Factory in order to pass reviews there. Sometimes leaf packages have to wait a long time due to low priority. If your Factory submission is stuck e.g. in legal review for too long and it affects 42.2 let us know please, so we can adjust priorities. - Maintainer review. Packages submitted by someone other than the Factory package maintainer(s) have a review set to the devel project/package to ask for permission by the Factory package maintainer. If there's no reply for too long, try adding a comment in the request to get some attention or talk to the project maintainers directly. cu Ludwig -- (o_ Ludwig Nussel //\ V_/_ http://www.suse.com/ SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

5 5

[opensuse-factory] new package for factory (TW) rubygem-libyui-rake
by Josef Reidinger 13 Sep '16

13 Sep '16

Hi dear factory audience, your yast team would like to introduce new package for factory - rubygem-libyui-rake. Package lives for certain time and we think it make sense to have it also in factory. What it is about? It is helper for checking and sending libyui ( graphic backend of YaST ) from git to proper target project ( e.g. from master to TW, from Leap branch to Leap releases or maintance, etc. ). It is very similar to already included rubygem-yast-rake. Who will maintain it? YaST Team. How often it will update? Releases are irregular, depending on found bugs and needs for new features. So road map, just on demand changes. Thanks for your time, Josef -- 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] Can I disable gpg check for rpms stored locally
by Roman Bysh 12 Sep '16

12 Sep '16

Hello, Every time I install my rpms (stored locally) using Yast, I get two pop up messages telling me they're broken and that my system can be broken if installed. Is there a setting where I can disable the pop up warnings for my local repository? -- Cheers! Roman IRC: 551368250 ============== -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

5 15

[opensuse-factory] New Tumbleweed snapshot 20160911 released!
by Dominique Leuenberger 12 Sep '16

12 Sep '16

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: cairo gawk (4.1.3 -> 4.1.4) gnome-clocks (3.20.1 -> 3.20.2) gnome-online-miners (3.20.0 -> 3.20.1) ibus-libpinyin (1.7.92 -> 1.8.0) libpinyin (1.5.92 -> 1.6.0) python-PyYAML (3.11 -> 3.12) python-Twisted (15.5.0 -> 16.4.0) systemd wayland-protocols (1.6 -> 1.7) === Details === ==== cairo ==== Subpackages: cairo-devel libcairo-gobject2 libcairo-script-interpreter2 libcairo2 libcairo2-32bit - Add cairo-xlib-double-free.patch to fix double free in _get_image_surface(); patch taken from upstream git (fdo#91967, boo#997189). ==== gawk ==== Version update (4.1.3 -> 4.1.4) * GNU awk 4.1.4: * z/OS support updated * debugger improvements * return value of system() enhanced to convey more information * two-way pipe handling improved * The -d option now allows -d- to print to standard output. * The DJGPP port is now officially deprecated. ==== gnome-clocks ==== Version update (3.20.1 -> 3.20.2) Subpackages: gnome-shell-search-provider-gnome-clocks - Update to version 3.20.2: + Misc bugfixes. + Updated translations. ==== gnome-online-miners ==== Version update (3.20.0 -> 3.20.1) - Update to version 3.20.1 (boo#997057): + gdata: - Query Google Drive in smaller chunks instead of getting everything all at once (bgo#770656). - Too many nfo:Equipment objects are needlessly created (bgo#770658). ==== ibus-libpinyin ==== Version update (1.7.92 -> 1.8.0) - Update to 1.8.0 * use libpinyin 1.6.0 ==== libpinyin ==== Version update (1.5.92 -> 1.6.0) Subpackages: libpinyin-data libpinyin9 - Update to 1.6.0 * fixes https://github.com/libpinyin/ibus-libpinyin/issues/56 * fixes pinyin_remember_user_input function * fixes pinyin_get_character_offset function * fixes compute_pronunciation_possibility function * fixes pinyin_get_character_offset function ==== python-PyYAML ==== Version update (3.11 -> 3.12) - Update to 3.12 - Adding an implicit resolver to a derived loader should not affect the base loader (fixes issue #57). - Uniform representation for OrderedDict? across different versions of Python (fixes issue #61). - Fixed comparison to None warning (closes issue #64). - fix Source URL ==== python-Twisted ==== Version update (15.5.0 -> 16.4.0) - Update to version 16.4.0 For the complete list of changes plese see https://github.com/twisted/twisted/blob/twisted-16.4.0/NEWS - Update to version 16.3.0 For the complete list of changes plese see https://github.com/twisted/twisted/blob/twisted-16.3.0/NEWS - Update to version 16.2.0 For the complete list of changes plese see https://github.com/twisted/twisted/blob/twisted-16.2.0/NEWS - Update to version 16.1.1 For the complete list of changes plese see https://github.com/twisted/twisted/blob/twisted-16.1.1/NEWS - Update to version 16.1.0 For the complete list of changes plese see https://github.com/twisted/twisted/blob/twisted-16.1.0/NEWS - Update to version 16.0.0 For the complete list of changes plese see https://github.com/twisted/twisted/blob/twisted-16.0.0/NEWS ==== systemd ==== Subpackages: libsystemd0 libsystemd0-32bit libudev-devel libudev1 libudev1-32bit systemd-32bit systemd-bash-completion systemd-logger systemd-sysvinit udev - Fix is-enabled check in systemd-sysv-install (bsc#997268) ==== wayland-protocols ==== Version update (1.6 -> 1.7) - Update to new upstream release 1.7 * Includes the new version of the unstable xdg-shell v6 protocol. * The protocol have gone through extensive changes to how surfaces are created and mapped, and also adds a few features that was not possible before: - A client can now specify a minimum and/or maximum window geometry. - A client can now in a declarative manner position popup surfaces so that they will be visible to the user, e.g. by having them placed within the shell work area. - A client can map a popup without a grab; this intended to be used for relatively positioned child surfaces that should not be automatically dismissed, such as tooltips, and user interface elements like "popovers" in GTK+. -- 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] leap 42.2 beta1 - no ipv6 address on start up?
by Per Jessen 12 Sep '16

12 Sep '16

On a fresh 42.2 install in an ipv6-only environment, my network interface does not get an address assigned on start-up. I have to do ifdown/ifup for that to work. -- Per Jessen, Zürich (23.0°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-factory+owner(a)opensuse.org

3 4