[jira] [Created] (IMPALA-6353) Crash in snappy decompressor

["Vuk Ercegovac (JIRA)" <ji...@apache.org>]

Vuk Ercegovac created IMPALA-6353:
-------------------------------------

             Summary: Crash in snappy decompressor
                 Key: IMPALA-6353
                 URL: https://issues.apache.org/jira/browse/IMPALA-6353
             Project: IMPALA
          Issue Type: Bug
          Components: Backend
    Affects Versions: Impala 2.10.0
            Reporter: Vuk Ercegovac


Crash in latest rhel exhaustive run:

From the impalad that crashed: 
F1224 10:25:38.853723 13490 decompress.cc:527] Check failed: input != nullptr

int64_t SnappyDecompressor::MaxOutputLen(int64_t input_len, const uint8_t* input) {
  DCHECK(input != nullptr);
  size_t result;
  if (!snappy::GetUncompressedLength(reinterpret_cast<const char*>(input),
          input_len, &result)) {
    return -1;
  }
  return result;
}

Backtrace:
#7  0x0000000001a41d18 in impala::SnappyDecompressor::MaxOutputLen (this=0xe6790c00, input_len=0, input=0x0) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/util/decompress.cc:527
#8  0x0000000001a41dec in impala::SnappyDecompressor::ProcessBlock (this=0xe6790c00, output_preallocated=true, input_length=0, input=0x0, output_length=0x7ff925231218, output=0x7ff925231418) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/util/decompress.cc:540
#9  0x0000000001a324bf in impala::Codec::ProcessBlock32 (this=0xe6790c00, output_preallocated=true, input_length=0, input=0x0, output_length=0x7ff9252313fc, output=0x7ff925231418) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/util/codec.cc:181
#10 0x0000000001c55576 in impala::BaseScalarColumnReader::InitDictionary (this=0x11ccc3600) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/parquet-column-readers.cc:1016
#11 0x0000000001bf4d86 in impala::HdfsParquetScanner::InitDictionaries (this=0xb1e01c0, column_readers=...) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/hdfs-parquet-scanner.cc:1704
#12 0x0000000001bec14b in impala::HdfsParquetScanner::NextRowGroup (this=0xb1e01c0) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/hdfs-parquet-scanner.cc:692
#13 0x0000000001bea957 in impala::HdfsParquetScanner::GetNextInternal (this=0xb1e01c0, row_batch=0xa2e55f0) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/hdfs-parquet-scanner.cc:488
#14 0x0000000001be8f56 in impala::HdfsParquetScanner::ProcessSplit (this=0xb1e01c0) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/hdfs-parquet-scanner.cc:404
#15 0x0000000001b73092 in impala::HdfsScanNode::ProcessSplit (this=0x88cbc00, filter_ctxs=..., expr_results_pool=0x7ff9252325b0, scan_range=0x15d358e0) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/hdfs-scan-node.cc:532
#16 0x0000000001b724bd in impala::HdfsScanNode::ScannerThread (this=0x88cbc00) at /data/jenkins/workspace/impala-asf-master-exhaustive-rhel7/repos/Impala/be/src/exec/hdfs-scan-node.cc:442
...

Query that was run at the time of the crash:
select NULL
select * from functional.decimal_tbl

Several of these were run around the same time:
Requesting prioritized load of table(s): default.jdbc_complex_type_test





--
This message was sent by Atlassian JIRA
(v6.4.14#64029)