[
https://issues.apache.org/jira/browse/IMPALA-12123?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17722745#comment-17722745
]
ASF subversion and git services commented on IMPALA-12123:
----------------------------------------------------------
Commit 10b130facb8c0f06d9a0f5e051728c9b825eab14 in impala's branch
refs/heads/master from Joe McDonnell
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=10b130fac ]
IMPALA-12123 (part 2): Fix offsets for HDFS caching with subranges
When using HDFS caching, subrange reads are copying memory
from the HDFS cache buffer into the output buffer. When
the HDFS cache buffer starts at an offset into the file,
the subrange offsets need to be adjusted to point to the
relative place within the cache buffer. This fixes the
calculation to use the relative locations.
Testing:
- Modified the caching tests in disk-io-mgr-test to also
test with reads at an offset
- In a local minicluster, creating a table based on
alltypes_tiny_pages.parquet, added HDFS caching for the table,
ran the SQLs in parquet-page-index-alltypes-tiny-pages.test,
and checked the HDFS caching reads in the profile.
Change-Id: Icd0ccc677b090ef3a75047defb4d7bda4279dac6
Reviewed-on: http://gerrit.cloudera.org:8080/19877
Tested-by: Impala Public Jenkins <[email protected]>
Reviewed-by: Joe McDonnell <[email protected]>
> SIGSEGV in ScanRange::ReadSubRanges() when using HDFS caching
> -------------------------------------------------------------
>
> Key: IMPALA-12123
> URL: https://issues.apache.org/jira/browse/IMPALA-12123
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Affects Versions: Impala 4.3.0
> Reporter: Joe McDonnell
> Assignee: Joe McDonnell
> Priority: Critical
> Fix For: Impala 4.3.0
>
>
> We have seen a crash where multiple executors hit this SIGSEGV simultaneously:
>
> {noformat}
> #0 0x00007f42a5112cb5 in ?? ()
> #1 0x0000000001742dab in impala::io::ScanRange::ReadSubRanges
> (this=this@entry=0x9d1bc940, queue=queue@entry=0x11c700a0,
> buffer_desc=buffer_desc@entry=0x6f73f2c0, eof=eof@entry=0x7f39bbddb727) at
> scan-range.cc:275
> #2 0x000000000174550b in impala::io::ScanRange::DoRead
> (this=this@entry=0x9d1bc940, queue=queue@entry=0x11c700a0, disk_id=7) at
> scan-range.cc:219
> #3 0x000000000173a0d6 in impala::io::DiskQueue::DiskThreadLoop
> (this=0x11c700a0, io_mgr=0x1273e8c0) at disk-io-mgr.cc:504
> #4 0x00000000014b0355 in boost::function0<void>::operator()
> (this=0x7f39bbddbb40) at
> ../../../toolchain/toolchain-packages-gcc7.5.0/boost-1.61.0-p2/include/boost/function/function_template.hpp:770
> #5 impala::Thread::SuperviseThread(std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > const&,
> std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>
> > const&, boost::function<void ()>, impala::ThreadDebugInfo const*,
> impala::Promise<long, (impala::PromiseMode)0>*) (name=..., category=...,
> functor=..., parent_thread_info=<optimized out>,
> thread_started=0x7ffea5eca350) at thread.cc:360
> #6 0x00000000014b171b in
> boost::_bi::list5<boost::_bi::value<std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > >,
> boost::_bi::value<std::__cxx11::basic_string<char, std::char_traits<char>,
> std::allocator<char> > >, boost::_bi::value<boost::function<void ()> >,
> boost::_bi::value<impala::ThreadDebugInfo*>,
> boost::_bi::value<impala::Promise<long, (impala::PromiseMode)0>*>
> >::operator()<void (*)(std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > const&,
> std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>
> > const&, boost::function<void ()>, impala::ThreadDebugInfo const*,
> impala::Promise<long, (impala::PromiseMode)0>*),
> boost::_bi::list0>(boost::_bi::type<void>, void
> (*&)(std::__cxx11::basic_string<char, std::char_traits<char>,
> std::allocator<char> > const&, std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > const&, boost::function<void
> ()>, impala::ThreadDebugInfo const*, impala::Promise<long,
> (impala::PromiseMode)0>*), boost::_bi::list0&, int) (
> a=<synthetic pointer>..., f=<error reading variable>, this=0x13b30800) at
> ../../../toolchain/toolchain-packages-gcc7.5.0/boost-1.61.0-p2/include/boost/bind/bind.hpp:531
> #7 boost::_bi::bind_t<void, void (*)(std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > const&,
> std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>
> > const&, boost::function<void ()>, impala::ThreadDebugInfo const*,
> impala::Promise<long, (impala::PromiseMode)0>*),
> boost::_bi::list5<boost::_bi::value<std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > >,
> boost::_bi::value<std::__cxx11::basic_string<char, std::char_traits<char>,
> std::allocator<char> > >, boost::_bi::value<boost::function<void ()> >,
> boost::_bi::value<impala::ThreadDebugInfo*>,
> boost::_bi::value<impala::Promise<long, (impala::PromiseMode)0>*> >
> >::operator()() (this=0x13b307f8)
> at
> ../../../toolchain/toolchain-packages-gcc7.5.0/boost-1.61.0-p2/include/boost/bind/bind.hpp:1222
> #8 boost::detail::thread_data<boost::_bi::bind_t<void, void
> (*)(std::__cxx11::basic_string<char, std::char_traits<char>,
> std::allocator<char> > const&, std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > const&, boost::function<void
> ()>, impala::ThreadDebugInfo const*, impala::Promise<long,
> (impala::PromiseMode)0>*),
> boost::_bi::list5<boost::_bi::value<std::__cxx11::basic_string<char,
> std::char_traits<char>, std::allocator<char> > >,
> boost::_bi::value<std::__cxx11::basic_string<char, std::char_traits<char>,
> std::allocator<char> > >, boost::_bi::value<boost::function<void ()> >,
> boost::_bi::value<impala::ThreadDebugInfo*>,
> boost::_bi::value<impala::Promise<long, (impala::PromiseMode)0>*> > >
> >::run() (this=0x13b30640)
> at
> ../../../toolchain/toolchain-packages-gcc7.5.0/boost-1.61.0-p2/include/boost/thread/detail/thread.hpp:116
> #9 0x0000000001caf602 in thread_proxy ()
> #10 0x00007f42a8420ea5 in ?? ()
> #11 0x0000000000000000 in ?? (){noformat}
> The error reported is:
>
>
> {noformat}
> Crash reason: SIGSEGV
> Crash address: 0x7f3903c7f438
> Process uptime: not available{noformat}
> We are working on finding details about the query that hit this.
>
> This corresponds to this line of code:
>
> {noformat}
> Status ScanRange::ReadSubRanges(
> DiskQueue* queue, BufferDescriptor* buffer_desc, bool* eof, FileReader*
> file_reader) {
> buffer_desc->len_ = 0;
> while (buffer_desc->len() < buffer_desc->buffer_len()
> && sub_range_pos_.index < sub_ranges_.size()) {
> SubRange& sub_range = sub_ranges_[sub_range_pos_.index];
> int64_t offset = sub_range.offset + sub_range_pos_.bytes_read;
> int64_t bytes_to_read = min(sub_range.length -sub_range_pos_.bytes_read,
> buffer_desc->buffer_len() - buffer_desc->len());
> if (cache_.data != nullptr) {
> memcpy(buffer_desc->buffer_ + buffer_desc->len(),
> cache_.data + offset, bytes_to_read); <<< HERE
> } else {{noformat}
> This is reading from HDFS caching.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
