Quanlong Huang created ORC-319:
----------------------------------
Summary: C++ reader crash after Lz4DecompressionStream thrown a
ParseError
Key: ORC-319
URL: https://issues.apache.org/jira/browse/ORC-319
Project: ORC
Issue Type: Bug
Components: C++
Affects Versions: 1.4.3
Reporter: Quanlong Huang
Attachments: decompress_crash.orc
The c++ reader crashes on reading a corrupt file (see attachments).
{code:java}
terminate called after throwing an instance of 'orc::ParseError*’
{code}
It crashes after throwing a ParseError exception from
orc::Lz4DecompressionStream::decompress
{code:java}
(gdb) bt
#0 0x00007ffff6a12c37 in __GI_raise (sig=sig@entry=6) at
../nptl/sysdeps/unix/sysv/linux/raise.c:56
#1 0x00007ffff6a16028 in __GI_abort () at abort.c:89
#2 0x00007ffff7321535 in __gnu_cxx::__verbose_terminate_handler() () from
/usr/lib/x86_64-linux-gnu/libstdc++.so.6
#3 0x00007ffff731f6d6 in ?? () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#4 0x00007ffff731f703 in std::terminate() () from
/usr/lib/x86_64-linux-gnu/libstdc++.so.6
#5 0x00007ffff731f922 in __cxa_throw () from
/usr/lib/x86_64-linux-gnu/libstdc++.so.6
#6 0x000000000089ca24 in orc::Lz4DecompressionStream::decompress
(this=0xf4db40,
input=0xf53f68
"\343`\026x\305)\305\307\301,\360\200I\242\221M\241\215Qc\033\223R,\a\217\020\067#\023\063\v+\033;\a'\027\267\024k|r~\216\001\204\062\204PF\020\312\030B\231@(S\be\006\241\314!\224\005\204\262\224b\003k7Pb\342`\006b\006
f\004b&(\237\005\210Y\201\230\r\210١\230\323`\033\223\025+\307\066\246",
length=270, output=0x7ffff7f60010
"`\026x\305)\305\307\301,\360\200I\242\221M\241",
maxOutputLength=262144) at /mnt/volume1/orc/c++/src/Compression.cc:882
#7 0x000000000089c3c1 in orc::BlockDecompressionStream::Next (this=0xf4db40,
data=0x7fffffffe0f8, size=0x7fffffffe0f4) at
/mnt/volume1/orc/c++/src/Compression.cc:722
#8 0x00000000008ac4ba in NextNonEmpty (size=0x7fffffffe0f4,
data=0x7fffffffe0f8, input=0xf4db40) at google/protobuf/io/coded_stream.cc:63
#9 google::protobuf::io::CodedInputStream::Refresh
(this=this@entry=0x7fffffffe140) at google/protobuf/io/coded_stream.cc:551
#10 0x00000000008aa06c in CodedInputStream (input=<optimized out>,
this=0x7fffffffe140) at ./google/protobuf/io/coded_stream.h:1183
#11 google::protobuf::MessageLite::ParseFromZeroCopyStream (this=0xf57d10,
input=<optimized out>) at google/protobuf/message_lite.cc:169
#12 0x0000000000834541 in orc::readFooter (stream=0xf51680,
buffer=@0x7fffffffe2b0: 0xf4f450, footerOffset=1429, ps=..., memoryPool=...) at
/mnt/volume1/orc/c++/src/Reader.cc:925
#13 0x0000000000834b57 in orc::createReader
(stream=std::unique_ptr<orc::InputStream> containing 0xf51680, options=...) at
/mnt/volume1/orc/c++/src/Reader.cc:978
#14 0x000000000082d28c in scanFile (out=..., filename=0x7fffffffe776
"decompress_crash.orc", batchSize=1024) at
/mnt/volume1/orc/tools/src/FileScan.cc:32
#15 0x000000000082d5a5 in main (argc=1, argv=0x7fffffffe510) at
/mnt/volume1/orc/tools/src/FileScan.cc:84
(gdb) f 6
#6 0x000000000089ca24 in orc::Lz4DecompressionStream::decompress
(this=0xf4db40,
input=0xf53f68
"\343`\026x\305)\305\307\301,\360\200I\242\221M\241\215Qc\033\223R,\a\217\020\067#\023\063\v+\033;\a'\027\267\024k|r~\216\001\204\062\204PF\020\312\030B\231@(S\be\006\241\314!\224\005\204\262\224b\003k7Pb\342`\006b\006
f\004b&(\237\005\210Y\201\230\r\210١\230\323`\033\223\025+\307\066\246",
length=270, output=0x7ffff7f60010
"`\026x\305)\305\307\301,\360\200I\242\221M\241",
maxOutputLength=262144) at /mnt/volume1/orc/c++/src/Compression.cc:882
882 throw new ParseError(getName() + " - failed to decompress");
(gdb) l
877 char *output,
878 size_t maxOutputLength) {
879 int result = LZ4_decompress_safe(input, output,
static_cast<int>(length),
880 static_cast<int>(maxOutputLength));
881 if (result < 0) {
882 throw new ParseError(getName() + " - failed to decompress");
883 }
884 return static_cast<uint64_t>(result);
885 }
886
{code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
