OceanBase服务启动报资源不够

OceanBase 技术问题
#1

【 使用环境 】 测试环境
【 OB or 其他组件 】
【 使用版本 】oceanbase-all-in-one-4.1.0.0-100120230323143519.el7.x86_64.tar.gz
【问题描述】启动服务报错。obd cluster start obtest
【复现路径】问题出现前后相关操作
【问题现象及影响】
[root@test03 conf]# obd cluster deploy obtest -c all-components.yaml
±-------------------------------------------------------------------------------------------+
| Packages |
±-------------±--------±-----------------------±-----------------------------------------+
| Repository | Version | Release | Md5 |
±-------------±--------±-----------------------±-----------------------------------------+
| oceanbase-ce | 4.1.0.0 | 100000192023032010.el7 | 8439ecf8db5e0649bd49671b41ea9e8c85756b63 |
| obproxy-ce | 4.1.0.0 | 7.el7 | 2a9d9bf67f179dcca2a8c9e7c77373d94e7e2abe |
| obagent | 1.3.0 | 22.el7 | d57fbb4962b2fbecb6282358c59295fdfba4d6ac |
| ocp-express | 1.0.0 | 100000432023032015.el7 | 42c6fc921063f24f9e1072d75bfa7f21f42146e3 |
±-------------±--------±-----------------------±-----------------------------------------+
Repository integrity check ok
Parameter check ok
Open ssh connection ok
Cluster status check ok
Initializes observer work home x
[ERROR] OBD-1002: Fail to init server1(172.31.0.15) home path: /opt/oceanbase-all-in-one is not empty.

Initializes obproxy work home ok
Initializes obagent work home x
[ERROR] OBD-1002: Fail to init server1(172.31.0.15) home path: /opt/oceanbase-all-in-one is not empty.

Initializes ocp-express work home x
[ERROR] OBD-1002: Fail to init 172.31.0.15 home path: /opt/oceanbase-all-in-one is not empty.

See https://www.oceanbase.com/product/ob-deployer/error-codes .
Trace ID: cbc8cc1c-df4f-11ed-998f-fa163e0a93c5
If you want to view detailed obd logs, please run: obd display-trace cbc8cc1c-df4f-11ed-998f-fa163e0a93c5
[root@test03 conf]#

#3

这个目录里是有其他文件吗？保持空目录试试

#4

刚才信息贴错了，信息应该是这个。

[root@test03 conf]# obd cluster start obtest
Get local repositories ok
Search plugins ok
Open ssh connection ok
Load cluster param plugin ok
Check before start observer x
[WARN] OBD-1007: (172.31.0.15) The recommended number of open files is 655350 (Current value: 65535)
[ERROR] OBD-2000: (172.31.0.15) not enough memory. (Available: 1.6G, Need: 3.0G)
[WARN] OBD-1012: (172.31.0.15) clog and data use the same disk (/)
[ERROR] OBD-2003: (172.31.0.15) / not enough disk space. (Avail: 355.6G, Need: 394.0G)

Check before start obproxy ok
Check before start obagent ok
Check before start ocp-express x
[ERROR] OBD-1009: 172.31.0.15 ocp-express need config: [‘jdbc_url’, ‘jdbc_password’, ‘jdbc_username’, ‘ob_cluster_id’, ‘root_sys_password’, ‘server_addresses’, ‘agent_username’, ‘agent_password’]
[ERROR] OBD-4300: 172.31.0.15: failed to query java version, you may not have java installed

See https://www.oceanbase.com/product/ob-deployer/error-codes .
Trace ID: ef1f1e68-df50-11ed-8535-fa163e0a93c5
If you want to view detailed obd logs, please run: obd display-trace ef1f1e68-df50-11ed-8535-fa163e0a93c5
[root@test03 conf]#

#5

all-components.yaml 这个配置文件发下

#6

1 看看你机器还有多少内存可用

java 装了吗 环境变量配了吗?

#7

您去 .obd/cluster 里面找您的安装的 配置文件 config.yaml 里面可以修改的 。 自己改下。

#8

发一下配置的yaml

#9

[root@test03 conf]# egrep -v ‘^#|^$|#’ all-components.yaml
oceanbase-ce:
servers:
- name: server1
ip: 172.19.33.2
- name: server2
ip: 172.19.33.3
- name: server3
ip: 172.19.33.4
global:
devname: eth0
system_memory: 30G
skip_proxy_sys_private_check: true
enable_strict_kernel_release: false
tenant_name: ocp
max_cpu: 1
memory_size: 2G
server1:
home_path: /root/observer
zone: zone1
server2:
home_path: /root/observer
zone: zone2
server3:
home_path: /root/observer
zone: zone3
obproxy-ce:
depends:
- oceanbase-ce
servers:
- 172.19.33.6
global:
home_path: /root/obproxy
enable_cluster_checkout: false
skip_proxy_sys_private_check: true
enable_strict_kernel_release: false
obagent:
depends:
- oceanbase-ce
servers:
- name: server1
ip: 172.19.33.2
- name: server2
ip: 172.19.33.3
- name: server3
ip: 172.19.33.4
global:
home_path: /root/obagent
ocp-express:
depeneds:
- oceanbase-ce
- obproxy-ce
- obagent
servers:
- 172.19.33.5
global:
home_path: /root/ocp-server
[root@test03 conf]#

#10

[root@test03 conf]# egrep -v ‘^#|^$|#’ all-components.yaml
oceanbase-ce:
servers:
- name: server1
ip: 172.31.0.15
global:
devname: eth0
system_memory: 30G
skip_proxy_sys_private_check: true
enable_strict_kernel_release: false
tenant_name: ocp
max_cpu: 1
memory_size: 2G
server1:
home_path: /opt/oceanbase-all-in-one/observer
data_dir: /opt/oceanbase-all-in-one/data
redo_dir: /opt/oceanbase-all-in-one/redo
zone: zone1
server2:
home_path: /root/observer
zone: zone2
server3:
home_path: /root/observer
zone: zone3
obproxy-ce:
depends:
- oceanbase-ce
servers:
- 172.31.0.15
global:
home_path: /opt/oceanbase-all-in-one/obproxy
enable_cluster_checkout: false
skip_proxy_sys_private_check: true
enable_strict_kernel_release: false
obagent:
depends:
- oceanbase-ce
servers:
- name: server1
ip: 172.31.0.15
global:
home_path: /opt/oceanbase-all-in-one/obagent
ocp-express:
depeneds:
- oceanbase-ce
- obproxy-ce
- obagent
servers:
- 172.31.0.15
global:
home_path: /opt/oceanbase-all-in-one/ocp-server
[root@test03 conf]#

#11

2个问题：
1）只安装OB和监控的最低配置
–因为ocp-express需要obagent和obproxy，本身占用资源有限，
主要是OB占用资源较大，最低要求2C8G SSD磁盘，容量需要根据需求安排，建议最低不少于50G，且数据盘和日志盘分开
ocpexpress 根据监控集群规模决定，少于10节点， 1C2G 磁盘 20G
obagent+obproxy 不低于2c 2G
2）推荐使用obd web白屏方式部署，可以选择最小占用，会自己规划占用资源，而且可以自定义资源参数，比较方便。
https://www.oceanbase.com/docs/common-oceanbase-database-cn-10000000001697442