[Fix](parquet-reader) Fix parquet reader crash in set_dict(). #40643
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
run buildall |
github-actions
bot
added
the
approved
|
PR approved by at least one committer and no changes requested. |
|
PR approved by anyone and no changes requested. |
wuwenchi
approved these changes
Sep 13, 2024
kaka11chen
added a commit
to kaka11chen/doris
that referenced
this pull request
Sep 20, 2024
…#40643) ## Proposed changes ### Issue ``` *** is nereids: 1 *** tablet id: 4 Abort at 1725864966 (unix time) try "date -d @1725864966" if you are using GNU date *** *** Set a breakpoint in static void __GI_abort() to debug *** PC: @ 0x7f007fb4090a04 *** SIGSEGV (address not mapped to object 0xa0fa868a41d6) received by PID 404737 (TID 274135 OR 0x7ece29df700) from PID 1755584205; stack trace: *** #0 __GI_raise #1 __GI_abort apache#2 sig_handler apache#3 _sigaction apache#4 JVM_handle_linux_signal apache#5 _sigaction apache#6 doris::vectorized::ByteArrayDictDecoder::set_dict(std::unique_ptr<unsigned char[], std::default_delete<unsigned char[]>> &&, int, unsigned long) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/byte_array_dict_decoder.cpp:41 apache#7 doris::vectorized::ColumnChunkReader::_decode_dict_page() at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_column_chunk_reader.cpp:258 apache#8 doris::vectorized::ColumnChunkReader::next_page() at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_column_chunk_reader.cpp:105 apache#9 doris::vectorized::ParquetColumnReader::_read_column_data(doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_column_reader.cpp:508 apache#10 doris::vectorized::ScalarColumnReader::_next_value(doris::vectorized::ICollumn*, unsigned long, unsigned long*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_column_reader.cpp:699 apache#11 doris::vectorized::RowGroupReader::_read_column_data(doris::vectorized::Block*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>> &, std::vector<doris::vectorized::ColumnSelectVector>*, unsigned long, unsigned long*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_group_reader.cpp:425 apache#12 doris::vectorized::RowGroupReader::get_next_block(doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_group_reader.cpp:311 apache#13 doris::vectorized::ParquetReader::get_next(doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/format/parquet/vparquet_reader.cpp:533 apache#14 doris::vectorized::VFileScanner::_get_next_reader_block(doris::RuntimeState*, doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/scan/vfile_scanner.cpp:368 apache#15 doris::vectorized::VFileScanner::_get_block_impl(doris::RuntimeState*, doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/scan/vfile_scanner.cpp:411 apache#16 doris::vectorized::VScanner::get_block(doris::RuntimeState*, doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/scan/vscanner.cpp:431 apache#17 doris::vectorized::VScanner::get_block(doris::RuntimeState*, doris::vectorized::Block*, bool*) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/scan/vscanner.cpp:96 apache#18 doris::vectorized::ScannerScheduler::submit(doris::vectorized::ScannerContext*, std::shared_ptr<doris::vectorized::ScanTask>) at /mnt/disk1/yy/git/enterprise-core/be/src/vec/exec/scan/scanner_context.cpp:96 apache#19 doris::Thread::supervise_thread(void*) at /mnt/disk1/yy/git/enterprise-core/be/src/util/thread.cpp:499 apache#20 start_thread apache#21 clone in /lib64/libc.so.6 ``` ### Solution It is not known why the parquet dictionary page will be null in this case, causing a crash. This PR adds defensive code to prevent the crash.
yiguolei
pushed a commit
that referenced
this pull request
Sep 21, 2024
## Proposed changes Backport #40643
morningman
pushed a commit
that referenced
this pull request
Sep 26, 2024
## Proposed changes Backport #40643
morningman
pushed a commit
to morningman/doris
that referenced
this pull request
Nov 21, 2024
…#41074) ## Proposed changes Backport apache#40643
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Proposed changes
Issue
Solution
It is not known why the parquet dictionary page will be null in this case, causing a crash. This PR adds defensive code to prevent the crash.