doris执行查询导致节点宕机

Viewed 75

be.out
image.png

*** Query id: fca065a384eb4917-93365c01a42d0bc1 ***
*** tablet id: 0 ***
*** Aborted at 1724141710 (unix time) try "date -d @1724141710" if you are using GNU date ***
*** Current BE git commitID: ae923f7094 ***
*** SIGSEGV address not mapped to object (@0x10) received by PID 3554003 (TID 3554351 OR 0xfffca8d38ac0) from PID 16; stack trace: ***
0# doris::signal::(anonymous namespace)::FailureSignalHandler(int, siginfo_t*, void*) at /root/yunyou/doris/be/src/common/signal_handler.h:417
1# os::Linux::chained_handler(int, siginfo_t*, void*) in /opt/software/jdk8/jre/lib/aarch64/server/libjvm.so

fe.audit.out
image.png
2024-08-20 16:15:11,585 [query] |Client=183.250.202.161:53233|User=root|Db=test|State=ERR|ErrorCode=1105|ErrorMessage=RpcException, msg: io.grpc.StatusRuntimeException: UNAVAILABLE: Network closed for unknown reason|Time(ms)=1178|ScanBytes=0|ScanRows=0|ReturnRows=0|StmtId=892|QueryId=fca065a384eb4917-93365c01a42d0bc1|IsQuery=true|isNereids=true|feIp=10.0.0.11|Stmt=/* ApplicationName=DBeaver Ultimate 22.3.0 - SQLEditor <Script-1.sql> */ SELECT * FROM hrp_index_data2019_5 ORDER BY the_date LIMIT 0,200|CpuTimeMS=0|SqlHash=e03be21d328ecd40015cf2f55335b66c|peakMemoryBytes=0|SqlDigest=|TraceId=|WorkloadGroup=|FuzzyVariables=

很简单的一个sql查询,数据量也不大就90W左右,be节点只有一个,本来是存活的,一执行这条sql就立马挂掉,找不到其它信息可以帮助处理了。
客户端执行报错截图
image.png

1 Answers

看着这个堆栈没打出来,不太好分析
帮忙确定几个问题:

  1. 这个是2.0的具体哪个版本
  2. hrp_index_data2019_5表的建表语句,show create table 看下具体的建表语句
  3. 确定下是否稳定复现,是否一执行这个query就挂
  4. 如果稳定复现的话麻烦开下coredump,拿个core 文件的
    BE 宕机处理最佳实践

堆栈信息不全,无法判断,建议可以的话升级到新版本进行测试,先close