Fix stacktraces on MacOS #59690

nikitamikhaylov · 2024-02-07T02:39:48Z

Changelog category (leave one):

Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Fix stack unwinding on MacOS. This closes #53653

robot-clickhouse · 2024-02-07T02:41:23Z

This is an automated comment for commit 2a79f56 with description of existing statuses. It's updated for the latest CI running

❌ Click here to open a full report in a separate page

Successful checks

Check name	Description	Status
AST fuzzer	Runs randomly generated queries to catch program errors. The build type is optionally given in parenthesis. If it fails, ask a maintainer for help	✅ success
ClickBench	Runs [ClickBench](https://github.com/ClickHouse/ClickBench/) with instant-attach table	✅ success
ClickHouse build check	Builds ClickHouse in various configurations for use in further steps. You have to fix the builds that fail. Build logs often has enough information to fix the error, but you might have to reproduce the failure locally. The cmake options can be found in the build log, grepping for cmake. Use these options and follow the general build process	✅ success
Compatibility check	Checks that clickhouse binary runs on distributions with old libc versions. If it fails, ask a maintainer for help	✅ success
Docker keeper image	The check to build and optionally push the mentioned image to docker hub	✅ success
Docker server image	The check to build and optionally push the mentioned image to docker hub	✅ success
Docs check	Builds and tests the documentation	✅ success
Fast test	Normally this is the first check that is ran for a PR. It builds ClickHouse and runs most of stateless functional tests, omitting some. If it fails, further checks are not started until it is fixed. Look at the report to see which tests fail, then reproduce the failure locally as described here	✅ success
Flaky tests	Checks if new added or modified tests are flaky by running them repeatedly, in parallel, with more randomization. Functional tests are run 100 times with address sanitizer, and additional randomization of thread scheduling. Integrational tests are run up to 10 times. If at least once a new test has failed, or was too long, this check will be red. We don't allow flaky tests, read the doc	✅ success
Install packages	Checks that the built packages are installable in a clear environment	✅ success
Integration tests	The integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests	✅ success
Mergeable Check	Checks if all other necessary checks are successful	✅ success
Performance Comparison	Measure changes in query performance. The performance test report is described in detail here. In square brackets are the optional part/total tests	✅ success
SQLTest	There's no description for the check yet, please add it to tests/ci/ci_config.py:CHECK_DESCRIPTIONS	✅ success
SQLancer	Fuzzing tests that detect logical bugs with SQLancer tool	✅ success
Stateful tests	Runs stateful functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc	✅ success
Stress test	Runs stateless functional tests concurrently from several clients to detect concurrency-related errors	✅ success
Style check	Runs a set of checks to keep the code style clean. If some of tests failed, see the related log from the report	✅ success
Unit tests	Runs the unit tests for different release types	✅ success
Upgrade check	Runs stress tests on server version from last release and then tries to upgrade it to the version from the PR. It checks if the new server can successfully startup without any errors, crashes or sanitizer asserts	✅ success

Check name	Description	Status
CI running	A meta-check that indicates the running CI. Normally, it's in success or pending state. The failed status indicates some problems with the PR	⏳ pending
Stateless tests	Runs stateless functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc	❌ failure

nikitamikhaylov · 2024-02-07T02:42:39Z

Previously ClickHouse binary natively compiled on MacOS was writing lot of errors like

libunwind: malformed __unwind_info at 0x112A73C54 bad second level page

whereas cross-compiled binary just had empty traces.
I looked into the LLDB source code and found out that it uses backtrace function from MacOS SDK (somewhat like libc or system library). ClickHouse already depends on it, so this change won't make anything any worse

❯ otool -L ./clickhouse
./clickhouse:
	/usr/lib/libiconv.2.dylib (compatibility version 7.0.0, current version 7.0.0)
	/System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 2202.0.0)
	/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1336.61.1)

With this patch I was able to get stacktraces like this:

❯ ./clickhouse local --stacktrace -q "SELECT throwIf(1)"
Code: 395. DB::Exception: Value passed to 'throwIf' function is non-zero: while executing 'FUNCTION throwIf(1 :: 0) -> throwIf(1) UInt8 : 1'. (FUNCTION_THROW_IF_VALUE_IS_NON_ZERO), Stack trace (when copying this message, always include the lines below):

0. std::exception::capture() @ 0x00000001041226a4 in clickhouse
1. std::exception::exception[abi:v15000]() @ 0x0000000104122670 in clickhouse
2. Poco::Exception::Exception(String const&, int) @ 0x000000011cd0d1a4 in clickhouse
3. DB::Exception::Exception(DB::Exception::MessageMasked&&, int, bool) @ 0x000000010c530cbc in clickhouse
4. DB::Exception::Exception(DB::Exception::MessageMasked&&, int, bool) @ 0x000000010c530e20 in clickhouse
5. DB::Exception::Exception(String&&, int, bool) @ 0x000000010410dab4 in clickhouse
6. DB::Exception::Exception<String>(int, FormatStringHelperImpl<std::type_identity<String>::type>, String&&) @ 0x000000010410d98c in clickhouse
7. COW<DB::IColumn>::immutable_ptr<DB::IColumn> DB::(anonymous namespace)::FunctionThrowIf::execute<char8_t>(DB::IColumn const*, std::optional<String> const&, std::optional<int> const&) const @ 0x000000010ad57388 in clickhouse
8. DB::(anonymous namespace)::FunctionThrowIf::executeImpl(std::vector<DB::ColumnWithTypeAndName, std::allocator<DB::ColumnWithTypeAndName>> const&, std::shared_ptr<DB::IDataType const> const&, unsigned long) const @ 0x000000010ad56218 in clickhouse
9. DB::FunctionToExecutableFunctionAdaptor::executeImpl(std::vector<DB::ColumnWithTypeAndName, std::allocator<DB::ColumnWithTypeAndName>> const&, std::shared_ptr<DB::IDataType const> const&, unsigned long) const @ 0x0000000104110bcc in clickhouse
10. DB::IExecutableFunction::executeWithoutLowCardinalityColumns(std::vector<DB::ColumnWithTypeAndName, std::allocator<DB::ColumnWithTypeAndName>> const&, std::shared_ptr<DB::IDataType const> const&, unsigned long, bool) const @ 0x000000011354bbd4 in clickhouse
11. DB::IExecutableFunction::executeWithoutSparseColumns(std::vector<DB::ColumnWithTypeAndName, std::allocator<DB::ColumnWithTypeAndName>> const&, std::shared_ptr<DB::IDataType const> const&, unsigned long, bool) const @ 0x000000011354cb28 in clickhouse
12. DB::IExecutableFunction::execute(std::vector<DB::ColumnWithTypeAndName, std::allocator<DB::ColumnWithTypeAndName>> const&, std::shared_ptr<DB::IDataType const> const&, unsigned long, bool) const @ 0x000000011354db08 in clickhouse
13. DB::executeAction(DB::ExpressionActions::Action const&, DB::(anonymous namespace)::ExecutionContext&, bool) @ 0x00000001165f4110 in clickhouse
14. DB::ExpressionActions::execute(DB::Block&, unsigned long&, bool) const @ 0x00000001165f3370 in clickhouse
15. DB::ExpressionTransform::transform(DB::Chunk&) @ 0x000000011a0f7508 in clickhouse
16. DB::ISimpleTransform::transform(DB::Chunk&, DB::Chunk&) @ 0x000000011005aa2c in clickhouse
17. DB::ISimpleTransform::work() @ 0x0000000119a13684 in clickhouse
18. DB::executeJob(DB::ExecutingGraph::Node*, DB::ReadProgressCallback*) @ 0x0000000119a54d8c in clickhouse
19. DB::ExecutionThreadContext::executeTask() @ 0x0000000119a54abc in clickhouse
20. DB::PipelineExecutor::executeStepImpl(unsigned long, std::atomic<bool>*) @ 0x0000000119a5e550 in clickhouse
21. DB::PipelineExecutor::executeSingleThread(unsigned long) @ 0x0000000119a5e9cc in clickhouse
22. DB::PipelineExecutor::executeImpl(unsigned long, bool) @ 0x0000000119a5d628 in clickhouse
23. DB::PipelineExecutor::execute(unsigned long, bool) @ 0x0000000119a5cdd0 in clickhouse
24. DB::threadFunction(DB::PullingAsyncPipelineExecutor::Data&, std::shared_ptr<DB::ThreadGroup>, unsigned long, bool) @ 0x0000000119a76548 in clickhouse
25. DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0::operator()() const @ 0x0000000119a763dc in clickhouse
26. decltype(std::declval<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&>()()) std::__invoke[abi:v15000]<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&) @ 0x0000000119a76338 in clickhouse
27. decltype(auto) std::__apply_tuple_impl[abi:v15000]<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&, std::tuple<>&>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&, std::tuple<>&, std::__tuple_indices<>) @ 0x0000000119a762fc in clickhouse
28. decltype(auto) std::apply[abi:v15000]<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&, std::tuple<>&>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&, std::tuple<>&) @ 0x0000000119a76200 in clickhouse
29. ThreadFromGlobalPoolImpl<true>::ThreadFromGlobalPoolImpl<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&&)::'lambda'()::operator()() @ 0x0000000119a7610c in clickhouse
30. decltype(std::declval<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0>()()) std::__invoke[abi:v15000]<ThreadFromGlobalPoolImpl<true>::ThreadFromGlobalPoolImpl<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&&)::'lambda'()&>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&&) @ 0x0000000119a76058 in clickhouse
31. void std::__invoke_void_return_wrapper<void, true>::__call<ThreadFromGlobalPoolImpl<true>::ThreadFromGlobalPoolImpl<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&&)::'lambda'()&>(ThreadFromGlobalPoolImpl<true>::ThreadFromGlobalPoolImpl<DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0>(DB::PullingAsyncPipelineExecutor::pull(DB::Chunk&, unsigned long long)::$_0&&)::'lambda'()&) @ 0x0000000119a76018 in clickhouse

contrib/libunwind-cmake/unwind-override.c

src/Common/StackTrace.cpp

alexey-milovidov · 2024-02-07T02:46:03Z

contrib/libunwind-cmake/unwind-override.c

@@ -1,6 +1,10 @@
 #include <libunwind.h>

+#if defined(__APPLE__)
+int backtrace(void ** buffer, int size);
+#else
 int backtrace(void ** buffer, int size)
 {
    return unw_backtrace(buffer, size);


Why exactly it doesn't work on Apple Macintosh OS X?

Why unw_backtrace doesn't work? I have no idea. I tried to use libunwind just from llvm's master - the result is the same. The only thing which bothers me - why LLDB can unwind stack while ClickHouse cannot. This was the motivation for this change.

It is prohibitive on Linux with glibc.
glibc uses an idiotic implementation: - it locates a .so at runtime and does dlopen.

It also does not work on Musl.

But I thought that it could be acceptable on Mac.

robot-clickhouse added pr-improvement Pull request with some product improvements submodule changed At least one submodule changed in this PR. labels Feb 7, 2024