Uploaded image for project: 'IMPALA'
  1. IMPALA
  2. IMPALA-1845

Codegen crash while loading module

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Critical
    • Resolution: Works for Me
    • Impala 2.2
    • Impala 2.2
    • None
    • None

    Description

      This happened while running tpch q3. This is the first time this happened (the query runs fine sometimes).

      #0  0x00007ff4287fce37 in __GI_raise (sig=sig@entry=6)
          at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
      56	../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
      (gdb) bt
      #0  0x00007ff4287fce37 in __GI_raise (sig=sig@entry=6)
          at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
      #1  0x00007ff4287fe528 in __GI_abort () at abort.c:89
      #2  0x00007ff4279decf5 in os::abort(bool) ()
         from /usr/lib/jvm/java-7-oracle/jre/lib/amd64/server/libjvm.so
      #3  0x00007ff427b604f7 in VMError::report_and_die() ()
         from /usr/lib/jvm/java-7-oracle/jre/lib/amd64/server/libjvm.so
      #4  0x00007ff4279e3c0f in JVM_handle_linux_signal ()
         from /usr/lib/jvm/java-7-oracle/jre/lib/amd64/server/libjvm.so
      #5  <signal handler called>
      #6  tcmalloc::CentralFreeList::FetchFromSpans (
          this=this@entry=0x2d25500 <tcmalloc::Static::central_cache_+7296>)
          at src/central_freelist.cc:298
      #7  0x00000000020b1997 in tcmalloc::CentralFreeList::RemoveRange (
          this=0x2d25500 <tcmalloc::Static::central_cache_+7296>, 
          start=start@entry=0x7ff3a9cb0688, end=end@entry=0x7ff3a9cb0690, N=32)
          at src/central_freelist.cc:269
      #8  0x00000000020acc22 in tcmalloc::ThreadCache::FetchFromCentralCache (
          this=this@entry=0xb619880, cl=6, byte_size=80) at src/thread_cache.cc:162
      #9  0x00000000020cb098 in Allocate (cl=<optimized out>, size=<optimized out>, 
          this=0xb619880) at src/thread_cache.h:341
      #10 do_malloc (size=<optimized out>) at src/tcmalloc.cc:1068
      #11 cpp_alloc (nothrow=false, size=80) at src/tcmalloc.cc:1354
      #12 tc_new (size=80) at src/tcmalloc.cc:1530
      #13 0x00000000016ddfbc in llvm::BitcodeReader::ParseFunctionBody(llvm::Function*) ()
      #14 0x00000000016e50e3 in llvm::BitcodeReader::Materialize(llvm::GlobalValue*, std::string*) ()
      #15 0x00000000016e5225 in llvm::BitcodeReader::MaterializeModule(llvm::Module*, std::string*) ()
      #16 0x0000000001f71089 in llvm::Module::MaterializeAllPermanently(std::string*) ()
      #17 0x00000000016e4b59 in llvm::ParseBitcodeFile(llvm::MemoryBuffer*, llvm::LLVMContext&, std::string*) ()
      #18 0x0000000000ee88b5 in impala::LlvmCodeGen::LoadModule (codegen=0x7961700, 
          file="/home/casey/Impala/llvm-ir/impala-sse.ll", module=0x7ff3a9cb11e8)
          at /home/casey/Impala/be/src/codegen/llvm-codegen.cc:150
      #19 0x0000000000ee851e in impala::LlvmCodeGen::LoadFromFile (pool=0xb462dc0, 
          file="/home/casey/Impala/llvm-ir/impala-sse.ll", 
          id="9540fd111faf15ad:74ea16fba0e293b9", codegen=0xc569090)
          at /home/casey/Impala/be/src/codegen/llvm-codegen.cc:126
      #20 0x0000000000ee8f70 in impala::LlvmCodeGen::LoadImpalaIR (pool=0xb462dc0, 
          id="9540fd111faf15ad:74ea16fba0e293b9", codegen_ret=0xc569090)
          at /home/casey/Impala/be/src/codegen/llvm-codegen.cc:189
      #21 0x0000000000fc4913 in impala::RuntimeState::CreateCodegen (this=0xc568e00)
          at /home/casey/Impala/be/src/runtime/runtime-state.cc:193
      #22 0x0000000000fc5f2d in impala::RuntimeState::GetCodegen (this=0xc568e00, 
          codegen=0x7ff3a9cb17d0, initialize=true)
          at /home/casey/Impala/be/src/runtime/runtime-state.cc:315
      #23 0x00000000015ebbf4 in impala::PartitionedAggregationNode::Prepare (this=0xb08c600, 
          state=0xc568e00) at /home/casey/Impala/be/src/exec/partitioned-aggregation-node.cc:102
      #24 0x00000000015347d5 in impala::ExecNode::Prepare (this=0x9ec3440, state=0xc568e00)
          at /home/casey/Impala/be/src/exec/exec-node.cc:149
      #25 0x000000000162383b in impala::TopNNode::Prepare (this=0x9ec3440, state=0xc568e00)
          at /home/casey/Impala/be/src/exec/topn-node.cc:56
      #26 0x00000000014fb54a in impala::PlanFragmentExecutor::Prepare (this=0x9a08210, 
          request=...) at /home/casey/Impala/be/src/runtime/plan-fragment-executor.cc:233
      #27 0x0000000001135740 in impala::FragmentMgr::FragmentExecState::Prepare (this=0x9a08000, 
          exec_params=...) at /home/casey/Impala/be/src/service/fragment-exec-state.cc:43
      #28 0x000000000112c78e in impala::FragmentMgr::ExecPlanFragment (this=0x94b66c0, 
          exec_params=...) at /home/casey/Impala/be/src/service/fragment-mgr.cc:51
      

      The query completed two times but crashed on the third:

      casey@casey-desktop:~/Impala/tests/query_test$ py.test test_mem_usage_scaling.py -k q3=================================== test session starts ====================================
      platform linux2 -- Python 2.7.8 -- py-1.4.24 -- pytest-2.6.2 -- /usr/bin/python
      plugins: xdist
      collected 124 items 
      
      test_mem_usage_scaling.py::TestTpchMemLimitError::test_low_mem_limit_q3[mem_limit: 100 | exec_option: {'disable_codegen': False, 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0, 'batch_size': 0, 'num_nodes': 0} | table_format: parquet/none] PASSED
      test_mem_usage_scaling.py::TestTpchMemLimitError::test_low_mem_limit_q3[mem_limit: 150 | exec_option: {'disable_codegen': False, 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0, 'batch_size': 0, 'num_nodes': 0} | table_format: parquet/none] PASSED
      test_mem_usage_scaling.py::TestTpchMemLimitError::test_low_mem_limit_q3[mem_limit: 180 | exec_option: {'disable_codegen': False, 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0, 'batch_size': 0, 'num_nodes': 0} | table_format: parquet/none] FAILEDTraceback (most recent call last):
        File "/usr/lib/python2.7/logging/__init__.py", line 880, in emit
          stream.write(fs % msg)
        File "/usr/local/lib/python2.7/dist-packages/_pytest/capture.py", line 234, in write
          self.buffer.write(obj)
      ValueError: I/O operation on closed file
      Logged from file impala_connection.py, line 150
      
      
      ========================================= FAILURES =========================================
       TestTpchMemLimitError.test_low_mem_limit_q3[mem_limit: 180 | exec_option: {'disable_codegen': False, 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0, 'batch_size': 0, 'num_nodes': 0} | table_format: parquet/none] 
      test_mem_usage_scaling.py:140: in test_low_mem_limit_q3
          self.low_memory_limit_test(vector, 'tpch-q3', self.MIN_MEM_FOR_TPCH['Q3']);
      test_mem_usage_scaling.py:131: in low_memory_limit_test
          assert TestTpchMemLimitError.EXPECTED_ERROR_MSG in str(e)
      E   assert 'Memory limit exceeded' in "ImpalaBeeswaxException:\n INNER EXCEPTION: <class 'thrift.transport.TTransport.TTransportException'>\n MESSAGE: TSocket read 0 bytes"
      E    +  where 'Memory limit exceeded' = TestTpchMemLimitError.EXPECTED_ERROR_MSG
      E    +  and   "ImpalaBeeswaxException:\n INNER EXCEPTION: <class 'thrift.transport.TTransport.TTransportException'>\n MESSAGE: TSocket read 0 bytes" = str(ImpalaBeeswaxException())
      ----------------------------------- Captured stderr call -----------------------------------
      -- executing against localhost:21000
      use tpch_parquet;
      
      SET batch_size=0;
      SET num_nodes=0;
      SET disable_codegen=False;
      SET abort_on_error=1;
      SET mem_limit=180m;
      SET exec_single_node_rows_threshold=0;
      MainThread: Query Name: 
      TPCH-Q3
      
      -- executing against localhost:21000
      select
        l_orderkey,
        sum(l_extendedprice * (1 - l_discount)) as revenue,
        o_orderdate,
        o_shippriority
      from
        customer,
        orders,
        lineitem
      where
        c_mktsegment = 'BUILDING'
        and c_custkey = o_custkey
        and l_orderkey = o_orderkey
        and o_orderdate < '1995-03-15'
        and l_shipdate > '1995-03-15'
      group by
        l_orderkey,
        o_orderdate,
        o_shippriority
      order by
        revenue desc,
        o_orderdate
      limit 10;
      
      ================================= short test summary info ==================================
      FAIL test_mem_usage_scaling.py::TestTpchMemLimitError::()::test_low_mem_limit_q3[mem_limit: 180 | exec_option: {'disable_codegen': False, 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0, 'batch_size': 0, 'num_nodes': 0} | table_format: parquet/none]
      !!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: stopping after 1 failures !!!!!!!!!!!!!!!!!!!!!!!!!!
      ============================== 114 tests deselected by '-kq3' ==============================
      ============= 1 failed, 2 passed, 114 deselected, 3 warnings in 288.67 seconds =============
      

      Attachments

        Activity

          People

            skye Skye Wanderman-Milne
            caseyc casey
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: