华为云使用OBD部署OCP失败

[Elm]

【 使用环境 】生产环境
【 OB or 其他组件 】OCP
【 使用版本 】4.3
【问题描述】华为云使用OBD部署OCP失败
【复现路径】
【附件及日志】
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
Package oceanbase-ce-4.2.1.7-107000162024060611.el7 is available.
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
Download obproxy-ce-4.2.3.0-3.el7.aarch64.rpm (117.83 M)
Download obproxy-ce-4.2.3.0-3.el7.aarch64.rpm (117.83 M)
Package obproxy-ce-4.2.3.0-3.el7 is available.
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
Download ocp-server-ce-4.3.0-20240617185317.el7.noarch.rpm (591.76 M)
Download ocp-server-ce-4.3.0-20240617185317.el7.noarch.rpm (591.76 M)
Package ocp-server-ce-4.3.0-20240617185317.el7 is available.
install obproxy-ce-4.2.3.0 for local ok
install ocp-server-ce-4.3.0 for local ok
±--------------------------------------------------------------------------------------------+
| Packages |
±--------------±--------±-----------------------±-----------------------------------------+
| Repository | Version | Release | Md5 |
±--------------±--------±-----------------------±-----------------------------------------+
| oceanbase-ce | 4.2.1.7 | 107000162024060611.el7 | 1a03d87ec62c1880415d1911beb10e35a872d416 |
| obproxy-ce | 4.2.3.0 | 3.el7 | cab6154cd013719765c938795d773b9c247bd653 |
| ocp-server-ce | 4.3.0 | 20240617185317.el7 | 3c446168804b6236851f40cb51cadaad5a35a0d3 |
±--------------±--------±-----------------------±-----------------------------------------+
Repository integrity check ok
Load param plugin ok
Open ssh connection ok
Generate ocp server configuration ok
Parameter check ok
Cluster status check ok
Initializes observer work home ok
Initializes obproxy work home ok
Initializes ocp-server-ce work home ok
Remote oceanbase-ce-4.2.1.7-107000162024060611.el7-1a03d87ec62c1880415d1911beb10e35a872d416 repository install ok
Remote oceanbase-ce-4.2.1.7-107000162024060611.el7-1a03d87ec62c1880415d1911beb10e35a872d416 repository lib check !!
Remote obproxy-ce-4.2.3.0-3.el7-cab6154cd013719765c938795d773b9c247bd653 repository install ok
Remote obproxy-ce-4.2.3.0-3.el7-cab6154cd013719765c938795d773b9c247bd653 repository lib check ok
Remote ocp-server-ce-4.3.0-20240617185317.el7-3c446168804b6236851f40cb51cadaad5a35a0d3 repository install ok
Remote ocp-server-ce-4.3.0-20240617185317.el7-3c446168804b6236851f40cb51cadaad5a35a0d3 repository lib check !!
Try to get lib-repository
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
[WARN] Use centos 7 remote mirror repository for ubuntu 22.04
Download openjdk-jre-1.8.0_322-b09.el7.aarch64.rpm (30.08 M)
Download openjdk-jre-1.8.0_322-b09.el7.aarch64.rpm (30.08 M)
Package openjdk-jre-1.8.0_322-b09.el7 is available.
install openjdk-jre-1.8.0_322 for local ok
Remote oceanbase-ce-libs-4.2.1.7-107000162024060611.nonlse.el7-aa7857eecc1e0de2caa6cddcf4938d55a8153be1 repository install ok
Remote oceanbase-ce-4.2.1.7-107000162024060611.el7-1a03d87ec62c1880415d1911beb10e35a872d416 repository lib check ok
Remote openjdk-jre-1.8.0_322-b09.el7-664b8eaeca384810f407b3925be853a0c6523b92 repository install ok
Remote ocp-server-ce-4.3.0-20240617185317.el7-3c446168804b6236851f40cb51cadaad5a35a0d3 repository lib check ok
data deployed
Search plugins ok
Load cluster param plugin ok
Check before start observer ok
[WARN] OBD-1007: (112.24.96.41) The recommended number of max user processes is 655350 (Current value: 514056)
[WARN] OBD-1007: (112.24.96.41) The recommended number of core file size is unlimited (Current value: 0)
[WARN] OBD-1007: (112.24.96.41) The recommended number of stack size is unlimited (Current value: 8192)
[WARN] OBD-1012: (112.24.96.41) clog and data use the same disk (/data)

Start observer ok
observer program health check ok
Connect to observer ok
Initialize oceanbase-ce x
[ERROR] 112.24.96.41 obshell failed

[ERROR] Cluster init failed
Search plugins ok
Load cluster param plugin ok
Check before start obproxy ok
Start obproxy ok
obproxy program health check

[Elm]

[Giant]

这个obshell不是那么重要，但是部署不成功，还真不让启动。我也遇到了

1. 我这增加资源后，就没问题了
2. 物理资源有限的情况下，您如果有多台机器，将资源使用 均衡分配下

查看obd的日志
[root@sharding16-ocp log]# ls
obd obd.2024-07-16 obd.2024-07-25 obd.2024-07-26 obd.2024-08-01
[root@sharding16-ocp log]# pwd
/root/.obd/log

[辞霜]

麻烦帮忙取一份 obd日志： 默认保存在安装obd的用户home路径： cd ~/.obd/log/

[Elm]

obd.log (353.7 KB)

[辞霜]

OBD-4360: 112.24.96.41: clockdiff有问题
报错主机无clockdiff 后 通过主机查看存在clockdiff么了，可以用commond -v clockdiff查看命令是否在/usr/bin/下

如果没有可以直接cp一个

[Giant]

检查下地址正常不？ 您说下您的部署架构吧

[Elm]

单机部署，使用OBD部署OB报错；使用OBD部署OCP报错。
华为云ECS：Ubuntu22.04.4 LTS
使用端口：62881,62882,62886

需求：部署单机OceanBase_v4.2.1.7，OMS，OCP在此ECS上

[Elm]

root@cn-east-6044-mpp-001:~# commond -v clockdiff
commond: command not found
root@cn-east-6044-mpp-001:~# clockdiff --version
clockdiff from iputils 20211215

[辞霜]

找一下clockdiff路径，估计是在/usr/sbin/下复制一份到/usr/bin/

[Elm]

commond -v clockdiff 这个命令拼写错误了，在/usr/bin/下

[辞霜]

搭建集群的资源规格设置多少。也有可能是设置太小导致的

[Elm]

准备使用OBD搭建OCP平台，然后用OCP搭建OB的。但是第一步时，没让选择资源大小。

[Giant]

这种地方

[Elm]

使用默认配置安装，依旧报错
[2024-08-15 06:01:32.552] [DEBUG] – exited code 2, error output:
[2024-08-15 06:01:32.552] [DEBUG] ls: cannot access ‘/root/data/oceanbase/.meta’: No such file or directory
[2024-08-15 06:01:32.554] [DEBUG]
[2024-08-15 06:01:32.554] [DEBUG] –
[2024-08-15 06:01:32.554] [DEBUG] – ls: cannot access ‘/root/data/oceanbase/.meta’: No such file or directory
[2024-08-15 06:01:32.555] [DEBUG]
[2024-08-15 06:01:32.555] [DEBUG] – root@112.24.96.41 execute: cat /root/data/oceanbase/run/obshell.pid
[2024-08-15 06:01:32.602] [DEBUG] – exited code 1, error output:
[2024-08-15 06:01:32.602] [DEBUG] cat: /root/data/oceanbase/run/obshell.pid: No such file or directory
[2024-08-15 06:01:32.603] [DEBUG]
[2024-08-15 06:01:32.603] [DEBUG] – root@112.24.96.41 execute: strings /root/data/oceanbase/etc/observer.conf.bin
[2024-08-15 06:01:32.657] [DEBUG] – exited code 1, error output:
[2024-08-15 06:01:32.657] [DEBUG] strings: ‘/root/data/oceanbase/etc/observer.conf.bin’: No such file
[2024-08-15 06:01:32.657] [DEBUG]
[2024-08-15 06:01:32.657] [DEBUG] –
[2024-08-15 06:01:32.657] [DEBUG] – strings: ‘/root/data/oceanbase/etc/observer.conf.bin’: No such file
[2024-08-15 06:01:32.658] [DEBUG]
[2024-08-15 06:01:32.658] [DEBUG] – root@112.24.96.41 execute: cat /root/data/oceanbase/run/obshell.pid
[2024-08-15 06:01:32.706] [DEBUG] – exited code 1, error output:
[2024-08-15 06:01:32.706] [DEBUG] cat: /root/data/oceanbase/run/obshell.pid: No such file or directory
[2024-08-15 06:01:32.706] [DEBUG]
[2024-08-15 06:01:32.707] [DEBUG] – root@112.24.96.41 set env OB_ROOT_PASSWORD to ‘‘Q9Z@KUH@p9Nn’’
[2024-08-15 06:01:32.707] [DEBUG] – start obshell: cd /root/data/oceanbase; /root/data/oceanbase/bin/obshell admin start --ip 112.24.96.41 --port 2886
[2024-08-15 06:01:32.707] [DEBUG] – root@112.24.96.41 execute: cd /root/data/oceanbase; /root/data/oceanbase/bin/obshell admin start --ip 112.24.96.41 --port 2886
[2024-08-15 06:01:37.863] [DEBUG] – exited code 30, error output:
[2024-08-15 06:01:37.863] [DEBUG] e[31m[FAILED]e[0m create tcp listerner failed: listen tcp 112.24.96.41:2886: bind: cannot assign requested address
[2024-08-15 06:01:37.864] [DEBUG]
[2024-08-15 06:01:37.864] [DEBUG] obshell server exited with code 21, please check obshell.log for more details
[2024-08-15 06:01:37.864] [DEBUG]
[2024-08-15 06:01:37.864] [ERROR] 112.24.96.41 obshell failed
[2024-08-15 06:01:37.864] [DEBUG] - sub bootstrap ref count to 0
[2024-08-15 06:01:37.864] [DEBUG] - export bootstrap
[2024-08-15 06:01:37.865] [DEBUG] - plugin oceanbase-ce-py_script_bootstrap-4.2.1.4 result: False
[2024-08-15 06:01:37.865] [INFO] [ERROR] 112.24.96.41 obshell failed
[2024-08-15 06:01:37.865] [INFO]
[2024-08-15 06:01:37.865] [ERROR] Cluster init failed

[李天波]

重新安装系统试试，同样问题重新部署系统就安装成功了

[辞霜]

obshell报错，看一下是不是已经存在obshell，帮忙拿取下obshell日志observer的log目录里有个log_obshell目录，里面的就是 obshell 的日志。