使用flink写入数据后后be无法启动

【详述】使用官方flink-connector的json模式写入数据到starrocks后,be节点全部挂了
【背景】仅数据写入
【业务影响】
【StarRocks版本】例如:2.5.3 avx2=off
【集群规模】3fe(2 follower+1 ob)+3be(fe与be混部)
【机器信息】CPU虚拟核/内存/网卡,例如:8c32G、千兆网卡
【联系方式】社区群13-小小小小鹏,谢谢
【附件】

  • fe.log/beINFO/相应截图
  • 慢查询:
    • Profile信息

    • 并行度:show variables like ‘%parallel_fragment_exec_instance_num%’;
      ±------------------------------------±------+
      | Variable_name | Value |
      ±------------------------------------±------+
      | parallel_fragment_exec_instance_num | 1 |
      ±------------------------------------±------+

    • pipeline是否开启:show variables like ‘%pipeline%’;
      ±--------------------------------±------+
      | Variable_name | Value |
      ±--------------------------------±------+
      | enable_pipeline_engine | true |
      | enable_pipeline_query_statistic | true |
      | pipeline_dop | 0 |
      | pipeline_profile_level | 1 |
      ±--------------------------------±------+

    • be节点cpu和内存使用率截图

  • 查询报错:
  • be crash
    • be.out 3个节点一样的报错信息,编译已经关闭了avx2
      start time: Tue Apr 11 20:21:05 CST 2023
      UNKNOWN RELEASE (build 020dde5)
      query_id:00000000-0000-0000-0000-000000000000, fragment_instance:00000000-0000-0000-0000-000000000000
      tracker:process consumption: 29176272
      tracker:query_pool consumption: 0
      tracker:load consumption: 0
      tracker:metadata consumption: 156681
      tracker:tablet_metadata consumption: 45581
      tracker:rowset_metadata consumption: 100808
      tracker:segment_metadata consumption: 1316
      tracker:column_metadata consumption: 8976
      tracker:tablet_schema consumption: 3325
      tracker:segment_zonemap consumption: 672
      tracker:short_key_index consumption: 0
      tracker:column_zonemap_index consumption: 1344
      tracker:ordinal_index consumption: 3888
      tracker:bitmap_index consumption: 0
      tracker:bloom_filter_index consumption: 0
      tracker:compaction consumption: 0
      tracker:schema_change consumption: 0
      tracker:column_pool consumption: 0
      tracker:page_cache consumption: 0
      tracker:update consumption: 29
      tracker:chunk_allocator consumption: 0
      tracker:clone consumption: 0
      tracker:consistency consumption: 0
      *** Aborted at 1681215666 (unix time) try “date -d @1681215666” if you are using GNU date ***
      PC: @ 0x5bce0dd ra_init
      *** SIGILL (@0x5bce0dd) received by PID 17054 (TID 0x7fab9cdf4700) from PID 96264413; stack trace: ***
      @ 0x5532762 google::(anonymous namespace)::FailureSignalHandler()
      @ 0x7fabb9cb15d0 (unknown)
      @ 0x5bce0dd ra_init
      @ 0x3faf4a0 starrocks::DelVector::_add_dels()
      @ 0x3fafc5c starrocks::DelVector::add_dels_as_new_version()
      @ 0x3e134e4 starrocks::TabletUpdates::_apply_rowset_commit()
      @ 0x3e151e2 starrocks::TabletUpdates::do_apply()
      @ 0x465b825 starrocks::ThreadPool::dispatch_thread()
      @ 0x46565ca starrocks::thread::supervise_thread()
      @ 0x7fabb9ca9dd5 start_thread
      @ 0x7fabb92c4ead __clone
      @ 0x0 (unknown)

没有avx2导致的吧

机器是没有avx2,我利用docker编译的,把build里面的avx2改为off了也不行。是不是不能用docker编译出来呢

具体参考 常见 Crash / BUG / 优化 查询