【 使用环境 】测试环境
【 OB or 其他组件 】 oceanbase-all-in-one-4.3.5_bp4_20250910.el8.aarch64.tar.gz
【 使用版本 】4.3.5
【问题描述】麒麟V10 + 鲲鹏916安装，obd demo报错，明明有内存，提示内存不足
【复现路径】问题出现前后相关操作
【附件及日志】推荐使用OceanBase敏捷诊断工具obdiag收集诊断信息,详情参见链接（右键跳转查看）：
1 个赞
obd版本是多少
麻烦提供一份obshell日志。看下obshell版本
observer的log目录里有个log_obshell目录，里面的就是 obshell 的日志。
1 个赞
新用户无法上传文件，我摘了obshell的日志贴出来：
2025-10-24T17:19:11.247 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:40] advanceTask: node 3 operator 1
2025-10-24T17:19:11.247 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:80] sub task 3 state 3
2025-10-24T17:19:11.247 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:53] ready Task num 0, isFinished false, isSucceed true
2025-10-24T17:19:11.365 INFO [55941] [F000000000000000] [executor/executor.go:259] finishing local task 3
2025-10-24T17:19:11.582 INFO [55941] [F000000000000000] [executor/executor.go:274] finish task 3 end
2025-10-24T17:19:11.744 INFO [55941] [f8b13f7ee688c2ae] [common/middleware.go:334] API response OK: [GET /api/v1/info, client=127.0.0.1, traceId=f8b13f7ee688c2ae, duration=1824, status=200, data=127.0.0.1:2886]
2025-10-24T17:19:11.750 INFO [55941] [8ab2dc1afa237f06] [common/middleware.go:241] API request: [GET /api/v1/info, client=127.0.0.1, traceId=8ab2dc1afa237f06, body={}]
2025-10-24T17:19:12.248 INFO [55941] [LS00000000000000] [scheduler/dag_handler.go:35] advance dag 1
2025-10-24T17:19:12.249 INFO [55941] [LS00000000000000] [scheduler/node_handler.go:25] advance node 3
2025-10-24T17:19:12.249 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:40] advanceTask: node 3 operator 1
2025-10-24T17:19:12.249 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:80] sub task 3 state 5
2025-10-24T17:19:12.249 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:53] ready Task num 0, isFinished true, isSucceed true
2025-10-24T17:19:13.510 INFO [55941] [LS00000000000000] [scheduler/dag_handler.go:35] advance dag 1
2025-10-24T17:19:13.510 INFO [55941] [LS00000000000000] [scheduler/node_handler.go:25] advance node 4
2025-10-24T17:19:13.605 INFO [55941] [8ab2dc1afa237f06] [common/middleware.go:334] API response OK: [GET /api/v1/info, client=127.0.0.1, traceId=8ab2dc1afa237f06, duration=1855, status=200, data=127.0.0.1:2886]
2025-10-24T17:19:13.610 INFO [55941] [7ff51d2ef57039a9] [common/middleware.go:241] API request: [GET /api/v1/secret, client=127.0.0.1, traceId=7ff51d2ef57039a9, body={}]
2025-10-24T17:19:13.610 INFO [55941] [7ff51d2ef57039a9] [common/middleware.go:334] API response OK: [GET /api/v1/secret, client=127.0.0.1, traceId=7ff51d2ef57039a9, duration=0, status=200, data=127.0.0.1:2886]
2025-10-24T17:19:13.629 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:40] advanceTask: node 4 operator 1
2025-10-24T17:19:13.630 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:80] sub task 4 state 1
2025-10-24T17:19:13.630 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:242] subtask 4 update executer agent 127.0.0.1:2886 to 127.0.0.1:2886
2025-10-24T17:19:13.704 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:53] ready Task num 1, isFinished false, isSucceed true
2025-10-24T17:19:13.705 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:56] ready sub task 4 operator 1
2025-10-24T17:19:13.705 INFO [55941] [F000000000000000] [executor/pool.go:65] add task 4 to ExecutorPool
2025-10-24T17:19:13.705 INFO [55941] [F000000000000000] [executor/pool.go:77] remove task 4 from ExecutorPool
2025-10-24T17:19:13.705 INFO [55941] [F000000000000000] [executor/executor.go:129] try to start task 4
2025-10-24T17:19:13.718 INFO [55941] [177fb7b7ed213c38] [common/middleware.go:241] API request: [GET /api/v1/task/dag/maintain/agent?show_details=True, client=127.0.0.1, traceId=177fb7b7ed213c38, body={}]
2025-10-24T17:19:13.719 INFO [55941] [177fb7b7ed213c38] [common/middleware.go:669] verfiy request: /api/v1/task/dag/maintain/agent?show_details=True
2025-10-24T17:19:13.855 INFO [55941] [F000000000000000] [executor/executor.go:135] start to task 4 execute
2025-10-24T17:19:13.855 INFO [55941] [F000000000000000] [executor/executor.go:279] try to execute task 4
2025-10-24T17:19:13.855 INFO [55941] [F000000000000000] [executor/executor.go:314] execute task 4, execute
2025-10-24T17:19:13.855 INFO [55941] [F000000000000000] [task/task.go:306] task 4 Convert master to cluster agent execute log: convert self to cluster agent
2025-10-24T17:19:14.051 INFO [55941] [177fb7b7ed213c38] [common/middleware.go:334] API response OK: [GET /api/v1/task/dag/maintain/agent?show_details=True, client=127.0.0.1, traceId=177fb7b7ed213c38, duration=333, status=200, data=&{GenericDTO:0x40008c05c0 DagDetail:0x40001509c0}]
2025-10-24T17:19:14.061 INFO [55941] [e2be1933e9a41386] [common/middleware.go:241] API request: [GET /api/v1/task/dag/22130706433028861?show_details=True, client=127.0.0.1, traceId=e2be1933e9a41386, body={}]
2025-10-24T17:19:14.061 INFO [55941] [e2be1933e9a41386] [common/middleware.go:669] verfiy request: /api/v1/task/dag/22130706433028861?show_details=True
2025-10-24T17:19:14.063 INFO [55941] [e2be1933e9a41386] [common/middleware.go:334] API response OK: [GET /api/v1/task/dag/22130706433028861?show_details=True, client=127.0.0.1, traceId=e2be1933e9a41386, duration=3, status=200, data=&{GenericDTO:0x40008c1f20 DagDetail:0x4000150fc0}]
2025-10-24T17:19:14.465 INFO [55941] [F000000000000000] [task/task.go:306] task 4 Convert master to cluster agent execute log: sync agent binary
2025-10-24T17:19:14.465 INFO [55941] [F000000000000000] [ob/ob_meta.go:70] try to upgrade binary
2025-10-24T17:19:14.645 INFO [55941] [F000000000000000] [engine/enter.go:54] cluster task engine starting …
2025-10-24T17:19:14.654 INFO [55941] [F000000000000000] [engine/enter.go:57] start cluster task engine
2025-10-24T17:19:14.654 INFO [55941] [F000000000000000] [engine/enter.go:79] cluster task engine started
2025-10-24T17:19:14.654 INFO [55941] [F000000000000000] [executor/synchronizer.go:54] synchronizer starting
2025-10-24T17:19:14.655 INFO [55941] [F000000000000000] [coordinator/coordinator.go:84] coordinator start
2025-10-24T17:19:14.655 INFO [55941] [F000000000000000] [coordinator/coordinator.go:167] coordinator initializing…
2025-10-24T17:19:14.664 INFO [55941] [F000000000000000] [coordinator/coordinator.go:311] set maintainer: 127.0.0.1:2886, life time: 662187.000000, last update time: 2025-10-17 01:22:47.664867556 +0800 CST m=-662156.855863369, expiration time: 2025-10-17 01:22:52.664867556 +0800 CST m=-662151.855863369
2025-10-24T17:19:14.664 INFO [55941] [F000000000000000] [coordinator/coordinator.go:333] coordinator identity change from FAULTY to MAINTAINER
2025-10-24T17:19:14.665 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 3.999337813s
2025-10-24T17:19:14.665 INFO [55941] [F000000000000000] [coordinator/synchronizer.go:138] agent synchronizer starting
2025-10-24T17:19:14.665 INFO [55941] [CS00000000000000] [runtime/asm_arm64.s:1223] scheduler starting
2025-10-24T17:19:14.669 INFO [55941] [F000000000000000] [coordinator/synchronizer.go:207] notice agent 127.0.0.1:2886 to synchronize all_agents
2025-10-24T17:19:14.670 INFO [55941] [F000000000000000] [agent/binary.go:126] Upgrading binary, chunk: 0
2025-10-24T17:19:14.706 INFO [55941] [LS00000000000000] [scheduler/dag_handler.go:35] advance dag 1
2025-10-24T17:19:14.706 INFO [55941] [LS00000000000000] [scheduler/node_handler.go:25] advance node 4
2025-10-24T17:19:14.706 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:40] advanceTask: node 4 operator 1
2025-10-24T17:19:14.706 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:80] sub task 4 state 3
2025-10-24T17:19:14.707 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:53] ready Task num 0, isFinished false, isSucceed true
2025-10-24T17:19:15.069 ERROR [55941] [F000000000000000] [coordinator/synchronizer.go:157] get all agents from ob failed fields: error=“The current database is useless”
2025-10-24T17:19:15.088 ERROR [55941] [F000000000000000] [ob/convert.go:111] upgrade binary failed fields:, error=“Error 4013 (HY001): No memory or reach tenant memory limit”
2025-10-24T17:19:15.088 ERROR [55941] [F000000000000000] [task/task.go:304] task 4 Convert master to cluster agent execute log: ERROR: [MySQL.Error]: MySQL error: Error 4013 (HY001): No memory or reach tenant memory limit
2025-10-24T17:19:15.088 WARN [55941] [F000000000000000] [executor/executor.go:137] task 4 execute error fields:, error=“Error 4013 (HY001): No memory or reach tenant memory limit”
2025-10-24T17:19:15.089 INFO [55941] [F000000000000000] [executor/executor.go:259] finishing local task 4
2025-10-24T17:19:15.150 INFO [55941] [F000000000000000] [executor/executor.go:274] finish task 4 end
2025-10-24T17:19:15.655 INFO [55941] [F000000000000000] [executor/synchronizer.go:59] synchronizer started
2025-10-24T17:19:15.657 INFO [55941] [F000000000000000] [coordinator/synchronizer.go:254] sync agent: 127.0.0.1:2886(CLUSTER AGENT) by coordinator
2025-10-24T17:19:16.037 INFO [55941] [LS00000000000000] [scheduler/dag_handler.go:35] advance dag 1
2025-10-24T17:19:16.037 INFO [55941] [LS00000000000000] [scheduler/node_handler.go:25] advance node 4
2025-10-24T17:19:16.037 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:40] advanceTask: node 4 operator 1
2025-10-24T17:19:16.037 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:80] sub task 4 state 4
2025-10-24T17:19:16.037 INFO [55941] [LS00000000000000] [scheduler/task_handler.go:53] ready Task num 0, isFinished true, isSucceed false
2025-10-24T17:19:16.070 ERROR [55941] [F000000000000000] [coordinator/synchronizer.go:157] get all agents from ob failed fields: error=“The current database is useless”
2025-10-24T17:19:17.073 INFO [55941] [ebed7ebec860a09d] [common/middleware.go:241] API request: [GET /api/v1/task/dag/22130706433028861?show_details=True, client=127.0.0.1, traceId=ebed7ebec860a09d, body={}]
2025-10-24T17:19:17.073 INFO [55941] [ebed7ebec860a09d] [common/middleware.go:669] verfiy request: /api/v1/task/dag/22130706433028861?show_details=True
2025-10-24T17:19:17.075 INFO [55941] [ebed7ebec860a09d] [common/middleware.go:334] API response OK: [GET /api/v1/task/dag/22130706433028861?show_details=True, client=127.0.0.1, traceId=ebed7ebec860a09d, duration=2, status=200, data=&{GenericDTO:0x4000854d20 DagDetail:0x4000150fc0}]
2025-10-24T17:19:18.665 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
2025-10-24T17:19:19.578 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 4s
2025-10-24T17:19:23.578 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
2025-10-24T17:19:25.291 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 4s
2025-10-24T17:19:29.292 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
2025-10-24T17:19:29.769 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 4s
2025-10-24T17:19:33.770 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
2025-10-24T17:19:34.272 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 4s
2025-10-24T17:19:38.273 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
2025-10-24T17:19:38.738 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 4s
2025-10-24T17:19:42.739 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
2025-10-24T17:19:43.214 INFO [55941] [F000000000000000] [coordinator/coordinator.go:122] coordinator wait: 4s
2025-10-24T17:19:47.215 INFO [55941] [F000000000000000] [coordinator/coordinator.go:157] renewal maintainer
会不会是内存参数设置的不对，太小了呢
下载的安装包没做任何改动，直接安装的，去哪可以改内存限制？我也想改下，是不是他原生的安装包里面的参数就有问题，但是同样的方式在x86平台上就能安装，同样是麒麟V10操作系统，换成ARM的就不行
No memory or reach tenant memory limit。
obshell也报错No memory or reach tenant memory limit。内存扩容到12G部署即可
1 个赞
您看下我第一次的截图，下面有free -h命令的输出，显示内存还有66G呀
cd ~/.obd/demo
里面的yaml设置了ob可使用内存，建议修改里面的memory_limit内存即可
1 个赞
直接vi yaml文件，然后执行obd redeploy cluster demo。
obd demo和obd deploy命令不一样
用这种方式有新的报错，内存free的还有40G：
[WARN] OBD-1011: (127.0.0.1) The recommended value of fs.aio-max-nr is 1048576 (Current value: 65536)
[ERROR] OBD-2010: (127.0.0.1): system_memory too large. system_memory must be less than memory_limit/memory_limit_percentage., system_memory should be less than 0.75 * memory_limit/memory_limit_percentage.
[WARN] OBD-1012: (127.0.0.1) clog and data use the same disk (/)
You can use the obd cluster init4env demo command to automatically configure system parameters
See https://www.oceanbase.com/product/ob-deployer/error-codes .
Trace ID: d318fc30-b300-11f0-af5b-e0cc7acbae63
If you want to view detailed obd logs, please run: obd display-trace d318fc30-b300-11f0-af5b-e0cc7acbae63
[root@knode1 demo]# free -h
total used free shared buff/cache available
Mem: 94Gi 15Gi 40Gi 72Mi 37Gi 77Gi
Swap: 4.0Gi 0B 4.0Gi
我把memory_limit 改为了1G，其他的没动，是不是还要改其他哪里？
我把内存显示加大到30G，就可以启动了，谢谢了，是不是默认的ARM的安装参数不对？
memory_limit： 30G，大于系统内存*0.75 应该就好了
并不是，你使用的obd版本是多少
3.6.0 和一键安装包一起下载安装的，
obd --version
OceanBase Deploy: 3.6.0
REVISION: b36013bb09a84516e56db51dba78a9d9096735e7
BUILD_BRANCH: HEAD
BUILD_TIME: Sep 04 2025 10:50:55OURCE
Copyright (C) 2025 OceanBase
License Apache 2.0: Apache version 2 or later https://www.apache.org/licenses/LICENSE-2.0.
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.