为了更快的定位您的问题，请提供以下信息，谢谢
【详述】starrocks3.5 BE宕机Core文件，61服务器，2026-07-04 22：17发生一次；60服务器，2026-07-07 23：17发生一次
【背景】没有做操作
【业务影响】be宕机写入任务会报错
【是否存算分离】否
【StarRocks版本】3.5.12
【集群规模】3fe（2 follower+2leader）+16be（fe与be分开部署）
【机器信息】64C/512G/万兆
【联系方式】社区群29-小欧，谢谢
【附件】
- be.out
- start time: Wed Mar 11 13:53:38 CST 2026, server uptime: 13:53:38 up 1628 days, 12 min, 1 user, load average: 0.08, 0.15, 0.10
- Run with JEMALLOC_CONF: 'percpu_arena:percpu,oversize_threshold:0,muzzy_decay_ms:5000,dirty_decay_ms:5000,metadata_thp:auto,backgroun
- d_thread:true,prof:true,prof_active:false’
- F20260707 23:17:43.121201 140160101123840 timer_thread.cpp:303] Check failed: version.load(butil::memory_order_relaxed) == id_version
- ** + 2 (2 vs. 5) **
- 3.5.13 RELEASE (build dca7c0a distro centos arch x86_64)
- query_id:00000000-0000-0000-0000-000000000000, fragment_instance:00000000-0000-0000-0000-000000000000, plan_node_id:-1
- [1783437463.126][thread: 140160101123840] je_mallctl execute purge success
- [1783437463.126][thread: 140160101123840] je_mallctl execute dontdump success
- ***** Aborted at 1783437472 (unix time) try “date -d @1783437472” if you are using GNU date *****
- PC: @ 0x7f7c12645387 __GI_raise
- ***** SIGABRT (@0x266c1) received by PID 157377 (TID 0x7f79910a0700) LWP(158823) from PID 157377; stack trace: *****
- ** @ 0x7f7c14ba120b __pthread_once_slow**
- ** @ 0xbe3e254 google::(anonymous namespace)::FailureSignalHandler(int, siginfo_t*, void*)**
- ** @ 0x7f7c14baa630 (/usr/lib64/libpthread-2.17.so+0xf62f)**
- ** @ 0x7f7c12645387 __GI_raise**
- ** @ 0x7f7c12646a78 __GI_abort**
- ** @ 0x4d2f5a3 starrocks::failure_function()**
- ** @ 0xbe3304a google::LogMessage::Fail()**
- ** @ 0xbe34a84 google::LogMessageFatal::~LogMessageFatal()**
- ** @ 0xc01ba22 bthread::TimerThread::Task::try_delete()**
- ** @ 0xc01c183 bthread::TimerThread::run()**
- ** @ 0xc01cdc0 bthread::TimerThread::run_this(void*)**
- ** @ 0x7f7c14ba2ea5 start_thread**
- ** @ 0x7f7c1270d96d __clone**
- start time: Wed Jul 8 00:02:09 CST 2026, server uptime: 00:02:09 up 1746 days, 10:21, 0 users, load average: 211.56, 1045.50, 128
- 4.37
- Run with JEMALLOC_CONF: 'percpu_arena:percpu,oversize_threshold:0,muzzy_decay_ms:5000,dirty_decay_ms:5000,metadata_thp:auto,backgroun
-
d_thread:true,prof:true,prof_active:false’
[New LWP 158065]
[New LWP 158569]
[New LWP 157853]
[New LWP 157845]
[New LWP 157873]
[New LWP 158056]
[New LWP 157871]
Core was generated by `/opt/StarRocks/be/lib/starrocks_be’.
Program terminated with signal 6, Aborted.
#0 0x00007f7c126d485d in ?? ()
######## be.WARNING ######## 的日志
F20260707 23:17:43.121201 140160101123840 timer_thread.cpp:303] Check failed: version.load(butil::memory_order_relaxed) == id_version + 2 (2 vs. 5)
W20260707 23:17:44.904889 139148559906560 heartbeat_server.cpp:100] Fail to handle heartbeat: Shut down in progress: BE is shutting down cached master info: TMasterInfo(network_address=TNetworkAddress(hostname=10.6.201.104, port=9020), cluster_id=298722942, epoch=6, token=, backend_ip=10.6.201.60, http_port=8030, heartbeat_flags=0, backend_id=62514, min_active_txn_id=13456172, run_mode=SHARED_NOTHING, disabled_disks=[], decommissioned_disks=[], encrypted=0, stop_regular_tablet_report=1, node_type=Backend) received master info: TMasterInfo(network_address=TNetworkAddress(hostname=10.6.201.104, port=9020), cluster_id=298722942, epoch=6, token=, backend_ip=10.6.201.60, http_port=8030, heartbeat_flags=0, backend_id=62514, min_active_txn_id=13456179, run_mode=SHARED_NOTHING, disabled_disks=[], decommissioned_disks=[], encrypted=0, stop_regular_tablet_report=1, node_type=Backend)
W20260707 23:17:49.905300 139148559906560 heartbeat_server.cpp:100] Fail to handle heartbeat: Shut down in progress: BE is shutting down cached master info: TMasterInfo(network_address=TNetworkAddress(hostname=10.6.201.104, port=9020), cluster_id=298722942, epoch=6, token=, backend_ip=10.6.201.60, http_port=8030, heartbeat_flags=0, backend_id=62514, min_active_txn_id=13456172, run_mode=SHARED_NOTHING, disabled_disks=[], decommissioned_disks=[], encrypted=0, stop_regular_tablet_report=1, node_type=Backend) received master info: TMasterInfo(network_address=TNetworkAddress(hostname=10.6.201.104, port=9020), cluster_id=298722942, epoch=6, token=, backend_ip=10.6.201.60, http_port=8030, heartbeat_flags=0, backend_id=62514, min_active_txn_id=13456183, run_mode=SHARED_NOTHING, disabled_disks=[], decommissioned_disks=[], encrypted=0, stop_regular_tablet_report=1, node_type=Backend)
W20260708 00:02:21.383314 139852250072832 fragment_context.cpp:200] [Driver] Canceled, query_id=3c7caddf-7a1d-11f1-8cb6-6805cae5bce0, instance_id=3c7caddf-7a1d-11f1-8cb6-6805cae5bce3, reason=transmit chunk rpc failed [dest_instance_id=3c7caddf-7a1d-11f1-8cb6-6805cae5bce3] [dest=10.6.201.60:8060] detail:brpc failed, error=Host is down, error_text=[E112]Not connected to 10.6.201.60:8060 yet, server_id=1836 [R1][E112]Not connected to 10.6.201.60:8060 yet, server_id=1836 [R2][E112]Not connected to 10.6.201.60:8060 yet, server_id=1836 [R3][E112]Not connected to 10.6.201.60:8060 yet, server_id=1836
系统负载指标情况：
