2024-12-04T19:43:04.265Z | <Slava Dubeyko> I a trying to build Ceph from the source locally on VM (Ubuntu 22.04). I am following to the guide: <https://docs.ceph.com/en/latest/start/beginners-guide/>. But I have some strange issue during running 'ninja -j3'. I have error:[18/2233] Performing configure step for 'rocksdb_ext'
-- The CXX compiler identification is GNU 11.4.0
-- The C compiler identification is GNU 11.4.0
-- The ASM compiler identification is GNU
-- Found assembler: /usr/bin/gcc-11
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/g++-11 - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /usr/bin/gcc-11 - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Found ZLIB: /usr/lib/x86_64-linux-gnu/libz.so (found version "1.2.11")
-- Found lz4: /usr/lib/x86_64-linux-gnu/liblz4.so
-- Performing Test HAVE_SSE42
-- Performing Test HAVE_SSE42 - Success
-- Performing Test BUILTIN_ATOMIC
-- Performing Test BUILTIN_ATOMIC - Success
-- Found uring: /home/slavad/CEPH/ceph/build/src/liburing/src
-- Enabling RTTI in all builds
-- Performing Test HAVE_FALLOCATE
-- Performing Test HAVE_FALLOCATE - Success
-- Performing Test HAVE_SYNC_FILE_RANGE_WRITE
-- Performing Test HAVE_SYNC_FILE_RANGE_WRITE - Success
-- Performing Test HAVE_PTHREAD_MUTEX_ADAPTIVE_NP
-- Performing Test HAVE_PTHREAD_MUTEX_ADAPTIVE_NP - Success
-- Looking for malloc_usable_size
-- Looking for malloc_usable_size - found
-- Looking for sched_getcpu
-- Looking for sched_getcpu - found
-- Looking for getauxval
-- Looking for getauxval - not found
-- Looking for F_FULLFSYNC
-- Looking for F_FULLFSYNC - not found
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- ROCKSDB_PLUGINS:
-- ROCKSDB PLUGINS TO BUILD
-- Found Git: /usr/bin/git (found version "2.34.1")
fatal: no tag exactly matches '9fa4990159853479a222244574ca41202e4c95c1'
-- JNI library is disabled
-- Configuring done
-- Generating done
-- Build files have been written to: /home/slavad/CEPH/ceph/build/src/rocksdb
[28/2233] Performing build step for 'rocksdb_ext'
FAILED: src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-build src/rocksdb/librocksdb.a /home/slavad/CEPH/ceph/build/src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-build /home/slavad/CEPH/ceph/build/src/rocksdb/librocksdb.a |
2024-12-04T19:43:07.357Z | <Slava Dubeyko> Is it known issue? |
2024-12-04T20:45:47.414Z | <yuriw> @Laura Flores @David Galloway maybe you know? |
2024-12-04T21:15:07.666Z | <David Galloway> `fatal: no tag exactly matches '9fa4990159853479a222244574ca41202e4c95c1'` |
2024-12-04T21:47:08.073Z | <Slava Dubeyko> And what is the problem here? How to solve the issue? Thanks. |
2024-12-04T21:51:31.006Z | <yuriw> that sha1 tag does not exit |
2024-12-04T21:52:36.564Z | <Slava Dubeyko> I follow this steps: <https://docs.ceph.com/en/latest/start/beginners-guide/> |
2024-12-04T21:52:50.223Z | <Slava Dubeyko> What I did wrong then? |
2024-12-04T21:53:42.387Z | <Laura Flores> Hmm I had the same "fatal" error but my compile proceeded. So I don't think it's that:
```-- Looking for malloc_usable_size - found
-- Looking for sched_getcpu
-- Looking for sched_getcpu - found
-- Looking for getauxval
-- Looking for getauxval - not found
-- Looking for F_FULLFSYNC
-- Looking for F_FULLFSYNC - not found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- ROCKSDB_PLUGINS:
-- ROCKSDB PLUGINS TO BUILD
-- Found Git: /usr/bin/git (found version "2.43.5")
fatal: no tag exactly matches '9fa4990159853479a222244574ca41202e4c95c1'
-- JNI library is disabled
-- Configuring done (1.7s)
-- Generating done (0.3s)
-- Build files have been written to: /home/lflores/ceph/build/src/rocksdb
[146/1372] Generating ../../bin/cephadm
cephadm/build.py: Python Version: 3.9.20
cephadm/build.py: Argument: dest='/home/lflores/ceph/build/bin/cephadm'
cephadm/build.py: Argument: source=None
cephadm/build.py: Argument: python=None
cephadm/build.py: Argument: version_vars=[('CEPH_GIT_VER', 'fda09f28ab6cca8c6323862c7fd904e52603c07e'), ('CEPH_GIT_NICE_VER', '19.3.0-6309-gfda09f28ab6'), ('CEPH_RELEASE', '19'), ('CEPH_RELEASE_NAME', 'squid'), ('CEPH_RELEASE_TYPE', 'dev')]
cephadm/build.py: Argument: pip_use_venv='auto'
cephadm/build.py: Argument: bundled_dependencies='pip'
cephadm/build.py: Source Dir: /home/lflores/ceph/src/cephadm
cephadm/build.py: Destination Path: /home/lflores/ceph/build/bin/cephadm
cephadm/build.py: Installing dependencies using pip
cephadm/build.py: Running cmd: /usr/bin/python3.9 -m venv --help
cephadm/build.py: Attempting to create a virtualenv
cephadm/build.py: Running cmd: /usr/bin/python3.9 -m venv /tmp/tmpqsirupo6.cephadm.build/deps/_venv_
cephadm/build.py: Running cmd: /tmp/tmpqsirupo6.cephadm.build/deps/_venv_/bin/python3.9 -m pip install -U pip
Requirement already satisfied: pip in /tmp/tmpqsirupo6.cephadm.build/deps/_venv_/lib/python3.9/site-packages (21.3.1)
Collecting pip
Using cached pip-24.3.1-py3-none-any.whl (1.8 MB)``` |
2024-12-04T21:54:48.061Z | <Laura Flores> Did you run this step:
```git submodule update --init --recursive --progress``` |
2024-12-04T21:55:00.068Z | <Laura Flores> Since rocksdb is a submodule it would be worth verifying this |
2024-12-04T21:55:17.489Z | <Slava Dubeyko> Yes, I did. |
2024-12-04T21:55:53.823Z | <Slava Dubeyko> I can try to rebuild from scratch. But I did this. |
2024-12-04T21:57:10.365Z | <Slava Dubeyko> Maybe, there are some environment's requirements? |
2024-12-04T21:57:14.190Z | <Laura Flores> Ok. I am building on centos 9 stream, so I'll try pulling an ubuntu 22.04 container to see if I can reproduce. But this doesn't obviously look distro specific |
2024-12-04T21:58:06.487Z | <Laura Flores> Oh actually can you try running `ninja vstart -j3` ?
`ninja -j3` compiles the whole project, but adding the vstart target compiles enough for a vstart cluster. |
2024-12-04T22:00:34.897Z | <Slava Dubeyko> Let me try. Thanks |
2024-12-04T22:06:47.387Z | <Slava Dubeyko> [167/1372] Performing build step for 'rocksdb_ext'
FAILED: src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-build src/rocksdb/librocksdb.a /home/slavad/CEPH/ceph/build/src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-build /home/slavad/CEPH/ceph/build/src/rocksdb/librocksdb.a
cd /home/slavad/CEPH/ceph/build/src/rocksdb && /usr/bin/cmake --build /home/slavad/CEPH/ceph/build/src/rocksdb --target rocksdb && /usr/bin/cmake -E touch /home/slavad/CEPH/ceph/build/src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-build
[1/324] Building CXX object CMakeFiles/rocksdb.dir/cache/cache_key.cc.o
[2/324] Building CXX object CMakeFiles/rocksdb.dir/cache/cache_entry_roles.cc.o
[3/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_contents.cc.o
[4/324] Building CXX object CMakeFiles/rocksdb.dir/cache/secondary_cache.cc.o
[5/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_file_meta.cc.o
[6/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_log_format.cc.o
[7/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_file_addition.cc.o
[8/324] Building CXX object CMakeFiles/rocksdb.dir/cache/cache_reservation_manager.cc.o
[9/324] Building CXX object CMakeFiles/rocksdb.dir/cache/cache.cc.o
[10/324] Building CXX object CMakeFiles/rocksdb.dir/cache/sharded_cache.cc.o
[11/324] Building CXX object CMakeFiles/rocksdb.dir/cache/compressed_secondary_cache.cc.o
[12/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_file_garbage.cc.o
[13/324] Building CXX object CMakeFiles/rocksdb.dir/cache/charged_cache.cc.o
[14/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_garbage_meter.cc.o
[15/324] Building CXX object CMakeFiles/rocksdb.dir/cache/lru_cache.cc.o
[16/324] Building CXX object CMakeFiles/rocksdb.dir/cache/clock_cache.cc.o
[17/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_log_sequential_reader.cc.o
[18/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_file_reader.cc.o
[19/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_log_writer.cc.o
[20/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_file_cache.cc.o
[21/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/prefetch_buffer_collection.cc.o
[22/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_source.cc.o
[23/324] Building CXX object CMakeFiles/rocksdb.dir/db/blob/blob_fetcher.cc.o
[24/324] Building CXX object CMakeFiles/rocksdb.dir/db/compaction/sst_partitioner.cc.o
[25/324] Building CXX object CMakeFiles/rocksdb.dir/db/c.cc.o
[26/324] Building CXX object CMakeFiles/rocksdb.dir/db/compaction/compaction_picker.cc.o
[27/324] Building CXX object CMakeFiles/rocksdb.dir/db/builder.cc.o
[28/324] Building CXX object CMakeFiles/rocksdb.dir/db/column_family.cc.o
FAILED: CMakeFiles/rocksdb.dir/db/column_family.cc.o
/usr/bin/g++-11 -DHAVE_PCLMUL -DHAVE_SSE42 -DLZ4 -DOS_LINUX -DROCKSDB_FALLOCATE_PRESENT -DROCKSDB_IOURING_PRESENT -DROCKSDB_LIB_IO_POSIX -DROCKSDB_MALLOC_USABLE_SIZE -DROCKSDB_NO_DYNAMIC_EXTENSION -DROCKSDB_PLATFORM_POSIX -DROCKSDB_PORTABLE -DROCKSDB_PTHREAD_ADAPTIVE_MUTEX -DROCKSDB_RANGESYNC_PRESENT -DROCKSDB_SCHED_GETCPU_PRESENT -DSNAPPY -DZLIB -I/home/slavad/CEPH/ceph/src/rocksdb -I/home/slavad/CEPH/ceph/src/rocksdb/include -isystem /home/slavad/CEPH/ceph/src/rocksdb/third-party/gtest-1.8.1/fused-src -isystem /home/slavad/CEPH/ceph/build/src/liburing/src/include -Wno-deprecated-copy -Wno-pessimizing-move -W -Wextra -Wall -pthread -Wsign-compare -Wshadow -Wno-unused-parameter -Wno-unused-variable -Woverloaded-virtual -Wnon-virtual-dtor -Wno-missing-field-initializers -Wno-strict-aliasing -Wno-invalid-offsetof -fno-builtin-memcmp -g -DROCKSDB_USE_RTTI -fPIC -MD -MT CMakeFiles/rocksdb.dir/db/column_family.cc.o -MF CMakeFiles/rocksdb.dir/db/column_family.cc.o.d -o CMakeFiles/rocksdb.dir/db/column_family.cc.o -c /home/slavad/CEPH/ceph/src/rocksdb/db/column_family.cc
g++-11: fatal error: Killed signal terminated program cc1plus
compilation terminated. |
2024-12-04T22:08:27.889Z | <Laura Flores> Ah, okay, `g++-11: fatal error: Killed signal terminated program cc1plus` indicates that your laptop ran out of memory to compile everything. This can happen since the ceph project is so large. Has anyone mentioned getting a sepia node to you? |
2024-12-04T22:09:17.845Z | <Laura Flores> We have some remote machines with much more memory that a lot of developers use to compile the project. This page explains how to set up the VPN: <https://wiki.sepia.ceph.com/doku.php?id=vpnaccess> |
2024-12-04T22:09:26.266Z | <Slava Dubeyko> Yeah, I know. I am simply trying to compile locally at first. OK. I see. |
2024-12-04T22:09:39.334Z | <Slava Dubeyko> Thanks |
2024-12-04T22:11:27.595Z | <Laura Flores> Of course! You were doing the right thing by adding the -j parameter. You can type `nproc` to see how many cores you have available, and increase that if there are more available. You can also compile targets for individual components, such as `ninja osd` to compile just the osd code. This won't let you run a vstart cluster, but it can help to verify that any code is compiling while you set up the sepia VPN. |
2024-12-04T22:13:49.932Z | <Slava Dubeyko> Thank you for the help. |