8.1.1-10 segfault on startup (after crash)

garthg

Looks like the segfault is generated from the analyze row count task. As soon as startup hit the AnalyzeRowCount [Command] TMTask: row count analyze - start it would crash.

This occurred after starting the db up after it had crashed due to an OOM error. We had to upgrade to 9.0.0-2 in order to get db to restart successfully.

uname: 3.10.0-514.6.2.el7.x86_64 #1 SMP Thu Feb 23 03:04:39 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

Backtrace Generated by Error
Signal: [0x000000000000000b] PID: [0x000000000003fe59] PC: [0x0000000003508459] FP: [0x00007fcafaff5950] SIGSEGV: SEGV_MAPERR SI_ADDR : [0x0000000000000000]
(ZN6Basics9Backtrace11DoBacktraceEiiPvS1+0x89b) [0x36b64bb]
(ZN6Basics20GlobalSignalHandlers14logFatalSignalEiPvS1+0xd2) [0x3703ea4]
(_ZN6BasicsL26sigHandler_withinATryCatchEiP7siginfoPv+0x233) [0x3705216]
(_ZN6BasicsL10sigHandlerEiP7siginfoPv+0xe) [0x3705329]
(__restore_rt+0x0) [0x7fcbdc0685e0]
(_ZN3CAT17LogicalStatistics18invalidateColStatsEv+0x2d) [0x3508459]
(_ZN17TMEnhancedArcTask20executeUninterpretedEPKcPSobRSt6vectorIySaIyEEb+0x2581) [0x1799331]
(_ZN3CMD6TMTask3runEPKcPSobSt6vectorIySaIyEEb+0x69) [0x17f64f5]
(_Z20do_analyze_row_countPKcRSob+0xd2) [0x16516fe]
(_ZN2TM29TMServiceTask_AnalyzeRowCount15runTaskInternalEb+0x1fa) [0x1bb99c2]
(_ZN4Util16TimerServiceTask7runTaskEb+0x81) [0x3807517]
(_ZN4Util20TimerServiceTaskList18TaskSchedulingInfo10threadShimEPv+0x2ac) [0x380b718]
(_ZN7Session13ThreadManager12launchThreadERKN5boost9function0IvEEPKc+0x41a) [0x3579c14]
(thread_proxy+0x2f) [0x400716f]
(start_thread+0xc5) [0x7fcbdc060e25]
(clone+0x6d) [0x7fcbdb98234d]