routine load 任务Running但是数据无法写入

Question

有5个routineload任务，4个单表写入1个多表写入，执行routine load全部OtherMsg显示如下：

2025-12-12 09:11:18:[CANCELLED]cancelled: [INTERNAL_ERROR]VNodeChannel[1764906464613-10025], load_id=4433dff5aacf140-d6cc4dbbf6dacabe, txn_id=3093706, node=xxxx:8060, open failed, err: [CANCELLED]PStatus: (xxxx)[CANCELLED]fail to offer request to the work pool, pool=FifoThreadPool(name=brpc_heavy, queue_size=10240/10240, active_thread=128/128, total_get_wait_time=2717782770784487, total_put_wait_time=465364197), host: xxxx

请问是任务消费不过来需要修改什么配置吗？跑了几个小时又会复现

阿渊@SelectDB (没回帖直接加我主页微信) · Answer

是3.0 的哪个版本？
调整下 brpc_light_work_pool_max_queue_size 大小试试
出现该问题的时候打个pstack 收集下信息的：pstack be_pid > be_ps.txt

Zane · Answer

routine load 错误信息补充

2025-12-16 17:23:48:[INTERNAL_ERROR]VNodeChannel[1764906466461-10008], load_id=c453b37c19d0417f-850d502824c9ac51, txn_id=5497638, node=xxxx:8060, open failed, err: [INTERNAL_ERROR]failed to open tablet writer, error=RPC call is timed out, error_text=[E1008]Reached timeout=60000ms @xxxx:8060, info=VNodeChannel[1764906466461-10008], load_id=c453b37c19d0417f-850d502824c9ac51, txn_id=5497638, node=xxxx:8060, host: xxxx

routine load 任务Running但是数据无法写入

2 Answers