Ceph : Issueshttps://tracker.ceph.com/https://tracker.ceph.com/favicon.ico2021-10-20T10:47:05ZCeph
Redmine rgw - Bug #52979 (Resolved): librgw_op_tp.so and librgw_rados_tp.so are included in librgw2 packa...https://tracker.ceph.com/issues/529792021-10-20T10:47:05ZNathan Cutlerncutler@suse.cz
<p>librgw_op_tp.so and librgw_rados_tp.so are included in librgw2 package, but have a different major version number than librgw.</p>
<p>This causes RPMLINT to complain:</p>
<pre>
ceph.log:[18094s] librgw2.x86_64: E: shlib-policy-name-error SONAME: librgw_op_tp.so.1, expected package suffix: 1
ceph.log:[18094s] librgw2.x86_64: E: shlib-policy-name-error SONAME: librgw_rados_tp.so.1, expected package suffix: 1
</pre>
<p>( This bug was first reported at <a class="external" href="https://bugzilla.opensuse.org/show_bug.cgi?id=1191751">https://bugzilla.opensuse.org/show_bug.cgi?id=1191751</a> )</p> Orchestrator - Bug #52064 (Resolved): octopus: cephadm bootstrap --container-init broken in Octopushttps://tracker.ceph.com/issues/520642021-08-05T08:46:08ZNathan Cutlerncutler@suse.cz
<p>In Octopus, when the user provides the "--container-init" option to "cephadm bootstrap", all containerized daemons in the cluster are supposed to get the "--init" option in their unit.run files, to ensure that an init process runs as PID 1 in the container. (Without this, the containerized daemons fail to generate coredumps when they crash.)</p>
<p>However, due to a bug in Octopus, when the "--container-init" option is provided to "cephadm bootstrap", only the daemons actually created by the bootstrap operation (which in some cases might be just 1 MON and 1 MGR) get the "--init" in their unit.run. Daemons created post-bootstrap using "ceph orch apply" do not get the "--init" goodness, giving rise to an undesirable situation where 1 MON and 1 MGR generate coredumps when they crash, but all other daemons in the cluster fail to do so.</p>
<p>Since "--container-init" is already deprecated in Octopus, I'm not sure if newer versions of Ceph (pacific, quincy, etc.) suffer from this bug.</p> Ceph - Bug #51968 (Resolved): rpm/luarocks: simplify conditional and support Leap 15.3https://tracker.ceph.com/issues/519682021-07-30T10:36:02ZNathan Cutlerncutler@suse.czCeph - Bug #51797 (Resolved): ceph.spec: after being eliminated, FIRST_ARG crept back inhttps://tracker.ceph.com/issues/517972021-07-22T12:07:46ZNathan Cutlerncutler@suse.cz
<p>the use of FIRST_ARG had already been eliminated by<br />926433f5d45e557c42f050b43798ba29dc495e02 but was re-introduced later by<br />9466d7098573dafcfede5e9c852374fbbd99f9e7</p> Ceph - Bug #51768 (Resolved): ceph.spec: drop use of DISABLE_RESTART_ON_UPDATE (SUSE specific)https://tracker.ceph.com/issues/517682021-07-21T15:44:48ZNathan Cutlerncutler@suse.cz
<p>The variable DISABLE_RESTART_ON_UPDATE is deprecated and use of %service_del_postun_without_restart macro should be preferred these days.</p> Ceph - Bug #51622 (Resolved): master FTBFS on SUSE ("Macro expanded in comment on line 136")https://tracker.ceph.com/issues/516222021-07-10T13:00:35ZNathan Cutlerncutler@suse.cz
<pre>
[ 149s] + exec rpmbuild -ba --define '_srcdefattr (-,root,root)' --nosignature --define '_build_create_debug 1' --define 'disturl obs://build.opensuse.org/filesystems:ceph:master:upstream/openSUSE_Tumbleweed/351d97d56d82f4f0d321d3ec79cb98ef-ceph' /home/abuild/rpmbuild/SOURCES/ceph.spec
[ 149s] warning: Macro expanded in comment on line 136: %limit_build in the place of smp_limit_mem_per_job, please
[ 149s]
[ 149s] error: line 136: Unknown tag: _threads="`/usr/bin/getconf _NPROCESSORS_ONLN`"
</pre> Ceph - Bug #51578 (Resolved): FTBFS with fmt-8.xhttps://tracker.ceph.com/issues/515782021-07-07T20:43:37ZNathan Cutlerncutler@suse.cz
<p>FTBFS with fmt-8.x on Fedora 35/rawhide</p> Ceph - Bug #50436 (Can't reproduce): master FTBFS with GCC11 on openSUSE Tumbleweedhttps://tracker.ceph.com/issues/504362021-04-20T10:38:30ZNathan Cutlerncutler@suse.cz
<p>In an openSUSE Tumbleweed environment, Ceph (master branch) fails to build from source when using GCC 11:</p>
<pre>
[ 3256s] In file included from /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-
suse-linux/boost/include/boost/beast/http/empty_body.hpp:15,
[ 3256s] from /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/src/lib
rbd/migration/HttpProcessorInterface.h:7,
[ 3256s] from /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/src/lib
rbd/migration/HttpClient.h:10,
[ 3256s] from /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/src/lib
rbd/migration/HttpClient.cc:4:
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/inclu
de/boost/beast/http/message.hpp: In instantiation of 'class boost::beast::http::header<false, b
oost::beast::http::basic_fields<std::allocator<char> > >':
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/inclu
de/boost/beast/http/message.hpp:494:7: required from 'class boost::beast::http::message<false
, boost::beast::http::empty_body, boost::beast::http::basic_fields<std::allocator<char> > >'
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/inclu
de/boost/beast/http/parser.hpp:61:55: required from 'class boost::beast::http::parser<false,
boost::beast::http::empty_body>'
[ 3256s] /usr/include/c++/11/type_traits:1199:4: required from 'struct std::__is_trivially_copy_constructible_impl<boost::beast::http::parser<false, boost::beast::http::empty_body>, true>'
[ 3256s] /usr/include/c++/11/type_traits:1204:12: required from 'struct std::is_trivially_copy_constructible<boost::beast::http::parser<false, boost::beast::http::empty_body> >'
[ 3256s] /usr/include/c++/11/type_traits:3127:43: required from 'constexpr const bool std::is_trivially_copy_constructible_v<boost::beast::http::parser<false, boost::beast::http::empty_body> >'
[ 3256s] /usr/include/c++/11/optional:663:11: required from 'class std::optional<boost::beast::http::parser<false, boost::beast::http::empty_body> >'
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/src/librbd/migration/HttpClient.cc:221:63: required from here
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/include/boost/beast/http/message.hpp:265:38: error: static assertion failed: Fields type requirements not met
[ 3256s] 265 | static_assert(is_fields<Fields>::value,
[ 3256s] | ^~~~~
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/include/boost/beast/http/message.hpp:265:38: note: 'std::integral_constant<bool, false>::value' evaluates to false
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/include/boost/beast/http/message.hpp: In instantiation of 'class boost::beast::http::header<true, boost::beast::http::basic_fields<std::allocator<char> > >':
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/include/boost/beast/http/message.hpp:494:7: required from 'class boost::beast::http::message<true, boost::beast::http::empty_body, boost::beast::http::basic_fields<std::allocator<char> > >'
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/src/librbd/migration/HttpClient.cc:800:6: required from here
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/include/boost/beast/http/message.hpp:61:38: error: static assertion failed: Fields type requirements not met
[ 3256s] 61 | static_assert(is_fields<Fields>::value,
[ 3256s] | ^~~~~
[ 3256s] /home/abuild/rpmbuild/BUILD/ceph-17.0.0-3329-g4f6adfd148/x86_64-suse-linux/boost/include/boost/beast/http/message.hpp:61:38: note: 'std::integral_constant<bool, false>::value' evaluates to false
[ 3259s] [ 83%] Building CXX object src/librbd/CMakeFiles/rbd_internal.dir/migration/HttpStream.cc.o
[ 3270s] [ 84%] Building CXX object src/rgw/CMakeFiles/rgw_common.dir/rgw_rest_sts.cc.o
[ 3271s] [ 84%] Building CXX object src/rgw/CMakeFiles/rgw_common.dir/rgw_perf_counters.cc.o
[ 3272s] make[2]: *** [src/librbd/CMakeFiles/rbd_internal.dir/build.make:1590: src/librbd/CMakeFiles/rbd_internal.dir/migration/HttpClient.cc.o] Error 1
</pre> Ceph - Bug #49678 (Resolved): %make_build causes OOM on low-memory builderhttps://tracker.ceph.com/issues/496782021-03-09T18:28:52ZNathan Cutlerncutler@suse.czCeph - Bug #49583 (Resolved): Master RPM builds in OBS run out of memory while attempting in-memo...https://tracker.ceph.com/issues/495832021-03-03T14:45:29ZNathan Cutlerncutler@suse.cz
<p>b50fc9e61c39e6f9544b67cb5cd49c67bf6dd02e introduced the use of "multi-threaded xz compression so we can build the compressed src rpm and binary rpms with smaller latency". This causes Ceph builds of the upstream master branch to fail in the OBS because OBS workers can theoretically run a high number of concurrent execution threads (typically 16), yet do not have enough memory to sustain all-out in-memory xz compression at the maximum number of threads. Since build workers in OBS have heterogeneous hardware, the maximum sustainable concurrency has to be calculated at build-time (in the %build section of the spec file). However, the number of threads for xz compression is set much higher up in the spec file.</p>
<p>Since this change - b50fc9e61c39e6f9544b67cb5cd49c67bf6dd02e - caused OBS builds to start failing 100% of the time, and it's not clear how to calculate the maximum sustainable number of xz compression threads (like we are able to do for %_smp_mflags in the %build section of the spec file), we can at least surround these two lines in a conditional so they don't get executed in OBS builds.</p> Ceph - Bug #49556 (Resolved): Master RPM builds in OBS failing due to OOM (regression)https://tracker.ceph.com/issues/495562021-03-01T17:01:35ZNathan Cutlerncutler@suse.cz
<p>43b441f9a3bc907c17d52385251001ffcd5d3ff9 removed a bunch of code that was protecting OBS builds from OOM by setting build parallelism on the basis of available memory instead of the number of CPU threads the build hardware can sustain.</p>
<p>This commit also introduced the use of the %_smb_build_ncpus macro, which is not available in any SUSE operating system except Tumbleweed (because only very recent versions of RPM define this macro).</p>
<p>Something will need to be done to restore the "good build behavior" we had the OBS before this regression.</p> Ceph - Bug #49550 (Resolved): Master RPM build tries to git clone external projecthttps://tracker.ceph.com/issues/495502021-03-01T11:04:13ZNathan Cutlerncutler@suse.cz
<p>Regression probably caused by a49d1dbb32e2436ff2836a85b2fa84418f0a5fff - <a class="external" href="https://github.com/ceph/ceph/pull/39539">https://github.com/ceph/ceph/pull/39539</a></p>
<p>In OBS, build workers do not have Internet connectivity. Build recently started failing with:</p>
<pre>
[ 63s] -- Ceph/.git directory not found, parsing /home/abuild/rpmbuild/BUILD/ceph-17.0.0-1275-g5e197a21e6/src/.git_version for CEPH_GIT_VER and CEPH_GIT_NICE_VER
[ 63s] -- Performing Test HAS_VTA
[ 63s] -- Performing Test HAS_VTA - Success
[ 63s] CMake Error at /usr/share/cmake/Modules/ExternalProject.cmake:2542 (message):
[ 63s] error: could not find git for clone of pmdk_ext
[ 63s] Call Stack (most recent call first):
[ 63s] /usr/share/cmake/Modules/ExternalProject.cmake:3430 (_ep_add_download_command)
[ 63s] cmake/modules/Buildpmem.cmake:17 (ExternalProject_Add)
[ 63s] src/CMakeLists.txt:513 (build_pmem)
[ 63s]
[ 63s]
[ 63s] -- Configuring incomplete, errors occurred!
[ 63s] See also "/home/abuild/rpmbuild/BUILD/ceph-17.0.0-1275-g5e197a21e6/build/CMakeFiles/CMakeOutput.log".
[ 63s] See also "/home/abuild/rpmbuild/BUILD/ceph-17.0.0-1275-g5e197a21e6/build/CMakeFiles/CMakeError.log".
[ 63s] error: Bad exit status from /var/tmp/rpm-tmp.P2zDRB (%build)
[ 63s]
[ 63s]
[ 63s] RPM build errors:
[ 63s] Bad exit status from /var/tmp/rpm-tmp.P2zDRB (%build)
</pre> mgr - Feature #49400 (Resolved): pg_autoscaler avoid scale-down until there is pressurehttps://tracker.ceph.com/issues/494002021-02-20T19:33:10ZNathan Cutlerncutler@suse.czCeph - Bug #49356 (New): Eliminate lua-luarocks as a dependency, not only in SLE but also in open...https://tracker.ceph.com/issues/493562021-02-18T15:31:02ZNathan Cutlerncutler@suse.cz
<p>openSUSE Factory (Tumbleweed) currently has three versions of luarocks:</p>
<pre>
sesdev@master:~> zypper se -s luarocks
Loading repository data...
Reading installed packages...
S | Name | Type | Version | Arch | Repository
---+----------------+---------+-----------+--------+------------------------
| lua51-luarocks | package | 3.3.1-1.3 | noarch | openSUSE-Tumbleweed-Oss
i+ | lua53-luarocks | package | 3.3.1-1.3 | noarch | openSUSE-Tumbleweed-Oss
| lua54-luarocks | package | 3.3.1-1.3 | noarch | openSUSE-Tumbleweed-Oss
</pre>
<p>None of these provide a sensible default package like "lua-luarocks" or "luarocks". The ceph.spec.in file would need to keep track of which exact version of lua is in which SUSE build target, and that is a non-trivial amount of work which someone would have to do on an ongoing basis.</p> devops - Bug #14906 (Resolved): redhat-lsb-core dependency was dropped, but is still neededhttps://tracker.ceph.com/issues/149062016-02-26T16:58:02ZNathan Cutlerncutler@suse.cz
<p>The redhat-lsb-core dependency was removed in <a class="external" href="https://github.com/ceph/ceph/pull/6363">https://github.com/ceph/ceph/pull/6363</a> but it appears to still be needed. David Disseldorp reports:</p>
<p>src/common/util.cc calls the lsb_release binary via shell in the collect_sys_info() code path:</p>
<pre>
133 static void lsb_release_parse(map<string, string> *m, CephContext *cct)
134 {
135 FILE *fp = popen("lsb_release -idrc", "r");
</pre>
<p>On my test system (which doesn't have lsb-release package installed), this results in:<br />Feb 24 12:11:41 harrison ceph-osd<sup><a href="#fn3287">3287</a></sup>: sh: lsb_release: command not found<br />Feb 24 12:11:41 harrison ceph-osd<sup><a href="#fn3287">3287</a></sup>: 2016-02-24 12:11:41.557873 7fb334a84700 -1 osd.0 583 lsb_release_parse - pclose failed: (0) Success</p>