【OBD-2002】按照提示修改后未解决

OceanBase 技术问题
#1

【使用环境】 测试环境
【OB or 其他组件】 OBD (OceanBase Deployer) 及 OceanBase 数据库
【使用版本】 oceanbase-all-in-one-4.3.5_bp3_20250808.el7.aarch64
【问题描述】
在 openEuler 22.03 (aarch64架构) 操作系统上，使用针对 EL7 的 all-in-one 安装包部署 OceanBase 集群时，启动失败，报错 OBD-2002。操作过程中收到警告 OBD-1012: (127.0.0.1) clog and data use the same disk (/)。已尝试调整数据目录、检查依赖和端口，问题依旧。

【复现路径】

  1. 操作系统：openEuler 22.03 LTS aarch64
  2. 下载安装包：oceanbase-all-in-one-4.3.5_bp3_20250808.el7.aarch64.tar.gz
  3. 解压并安装 OBD。
  4. 使用 obd cluster deploy 命令部署名为 oceanbase-jamy 的集群。
  5. 执行 obd cluster start oceanbase-jamy 启动集群。
  6. 命令执行失败，返回 ERROR OBD-2002，并提示查看 trace。同时有 WARN OBD-1012。

【附件及日志】
已运行 obd display-trace 496214d0-8156-11f0-a761-707990b8c776 获取详细日志，但建议官方人员指导或提供更全面的日志收集方法（如 obdiag），因为当前环境部署未成功，obdiag 可能无法直接使用。关键错误信息如下：

[WARN] OBD-1012: (127.0.0.1) clog and data use the same disk (/) cluster scenario: htap Start observer x [ERROR] OBD-2002: Failed to start 127.0.0.1 observer: See https://www.oceanbase.com/product/ob-deployer/error-codes . Trace ID: 496214d0-8156-11f0-a761-707990b8c776

【备注】

  1. 怀疑可能与 openEuler 22.03 和 RPM 包针对的 EL7 版本之间的兼容性有关。
  2. 已尝试的解决方案（无效）：
    修改memory_limit 和 system_memory:
    image
  3. 服务器版本：
    [root@localhost oceanbase]# uname -a
    Linux localhost.localdomain 5.10.0-60.139.0.166.oe2203.aarch64 #1 SMP Thu May 30 05:18:35 UTC 2024 aarch64 aarch64 aarch64 GNU/Linux
    [root@localhost oceanbase]# cat /etc/os-release
    NAME=“openEuler”
    VERSION=“22.03 LTS”
    ID=“openEuler”
    VERSION_ID=“22.03”
    PRETTY_NAME=“openEuler 22.03 LTS”
    ANSI_COLOR=“0;31”
#4

需要提供一份obd的日志，system memory建议设置为3G
~/.obd/log

1 个赞
#5

你好，修改参数后仍然启动失败，新用户不能上传附件，obd日志如下：
[WARNING] OBD-1007: (127.0.0.1) The recommended number of open files is 655350 (Current value: 65535)
[WARNING] OBD-1012: (127.0.0.1) clog and data use the same disk (/)
DEBUG – local execute: cd /root/oceanbase-jamy; /root/oceanbase-jamy/bin/observer -r ‘127.0.0.1:2889:2888’ -p 2888 -P 2889 -z ‘zone1’ -n ‘oceanbase-jamy’ -c 1756089503 -d ‘/root/oceanbase-jamy/store’ -i ‘lo’ -o __min_full_resource_pool_memory=1073741824,memory_limit=‘12G’,system_memory=‘3G’,enable_syslog_recycle=True,enable_syslog_wf=False,max_syslog_file_count=4,cpu_count=2,datafile_size=‘5G’,log_disk_size=‘5G’,large_query_threshold=‘600s’,enable_record_trace_log=False
DEBUG – exited code -4, error output:
[ERROR] OBD-2002: Failed to start 127.0.0.1 observer:

#6

[2025-08-25 10:38:34.080] [DEBUG] – local execute: ls /root/oceanbase-jamy/store/clog/tenant_1/
[2025-08-25 10:38:34.088] [DEBUG] – exited code 2, error output:
[2025-08-25 10:38:34.088] [DEBUG] ls: cannot access ‘/root/oceanbase-jamy/store/clog/tenant_1/’: No suc h file or directory
[2025-08-25 10:38:34.089] [DEBUG]
[2025-08-25 10:38:34.089] [DEBUG] – local execute: cat /root/oceanbase-jamy/run/observer.pid
[2025-08-25 10:38:34.096] [DEBUG] – exited code 1, error output:
[2025-08-25 10:38:34.096] [DEBUG] cat: /root/oceanbase-jamy/run/observer.pid: No such file or directory
[2025-08-25 10:38:34.096] [DEBUG]

#7

启动ob时候失败，需要提供一下observer日志看看是什么原因导致。
cd ~/xxxx/oceanbase/log

#8

你好，没有找到这个日志，log文件夹里没有任何文件
image

#9

cd ~/oceanbase-jamy/oceanbase/log

#10

没有这个文件夹
image

#11

直接cd log

#12

image

#13

重新部署一下看下目录生成日志么

#14

这是我重新部署的步骤：

obd cluster destroy oceanbase-jamy
obd cluster deploy oceanbase-jamy -c /home/oceanbase/oceanbase-jamy.yaml
obd cluster start oceanbase-jamy

运行后报错如下：

[WARN] OBD-1007: (127.0.0.1) The recommended number of stack size is unlimited (Current value: 8192)
[WARN] OBD-1012: (127.0.0.1) clog and data use the same disk (/)
cluster scenario: htap
Start observer x
[ERROR] OBD-2002: Failed to start 127.0.0.1 observer:
See https://www.oceanbase.com/product/ob-deployer/error-codes .
Trace ID: f18070f6-817b-11f0-9d90-707990b8c776
If you want to view detailed obd logs, please run: obd display-trace f18070f6-817b-11f0-9d90-707990b8c776

oceanbase-jamy.yaml 内容如下：

oceanbase-ce:
  servers:
  - 127.0.0.1
  global:
    home_path: /root/oceanbase-jamy
    devname: lo
    mysql_port: 2888
    rpc_port: 2889
    zone: zone1
    memory_limit: 12G
    system_memory: 3G
    enable_syslog_recycle: true
    enable_syslog_wf: false
    max_syslog_file_count: 4
    production_mode: false
    __min_full_resource_pool_memory: 1073741824
    cpu_count: 2
    datafile_size: 5G
    log_disk_size: 5G

还是没有生成日志：
image

#15

log_disk_size这个参数调大点 40G

#16

调成40G后报错

[ERROR] OBD-2003: (127.0.0.1) / not enough disk space. (Avail: 48G, Need: 49G), Please reduce the `datafile_size` or `datafile_disk_percentage`

后来我调成30G，还是之前的报错

[ERROR] OBD-2002: Failed to start 127.0.0.1 observer:

log文件夹里还是没有日志生成

#17

使用redeploy重新部署试一下

#18

很遗憾还是报错[ERROR] OBD-2002，log目录还是空的。
这是我最后yaml文件的参数：

oceanbase-ce:
  servers:
  - 127.0.0.1
  global:
    home_path: /root/oceanbase-jamy
    devname: lo
    mysql_port: 2888
    rpc_port: 2889
    zone: zone1
    memory_limit: 10G
    system_memory: 3G
    enable_syslog_recycle: true
    enable_syslog_wf: false
    max_syslog_file_count: 4
    production_mode: false
    __min_full_resource_pool_memory: 1073741824
    cpu_count: 2
    datafile_size: 5G
    log_disk_size: 30G
#19

obd的日志发一份看看
~/.obd/log

#20

还是不能上传附件，这是最后50行的输出：

[root@localhost log]# tail -n 50 obd
[2025-08-25 15:05:43.309] [e55967dc-8181-11f0-81e6-707990b8c776] [INFO] cluster scenario: htap
[2025-08-25 15:05:43.309] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- local execute: ls /root/oceanbase-jamy/store/clog/tenant_1/
[2025-08-25 15:05:43.316] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- exited code 2, error output:
[2025-08-25 15:05:43.317] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] ls: cannot access '/root/oceanbase-jamy/store/clog/tenant_1/': No such file or directory
[2025-08-25 15:05:43.317] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG]
[2025-08-25 15:05:43.317] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- local execute: cat /root/oceanbase-jamy/run/observer.pid
[2025-08-25 15:05:43.324] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- exited code 1, error output:
[2025-08-25 15:05:43.324] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] cat: /root/oceanbase-jamy/run/observer.pid: No such file or directory
[2025-08-25 15:05:43.324] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG]
[2025-08-25 15:05:43.324] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- 127.0.0.1 start command construction
[2025-08-25 15:05:43.324] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- update large_query_threshold to 600s because of scenario
[2025-08-25 15:05:43.325] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- update enable_record_trace_log to False because of scenario
[2025-08-25 15:05:43.325] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- update enable_syslog_recycle to 1 because of scenario
[2025-08-25 15:05:43.325] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- update max_syslog_file_count to 300 because of scenario
[2025-08-25 15:05:43.325] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - sub start_pre ref count to 0
[2025-08-25 15:05:43.326] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - export start_pre
[2025-08-25 15:05:43.326] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - plugin oceanbase-ce-py_script_start_pre-4.2.5.0 result: True
[2025-08-25 15:05:43.326] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - Searching start plugin for components ...
[2025-08-25 15:05:43.326] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - Searching start plugin for oceanbase-ce-4.3.5.3-103000092025080818.el7-51410153665c32c5e0c66a4cb1cbe94648da12fb
[2025-08-25 15:05:43.328] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - Found for oceanbase-ce-py_script_start-3.1.0 for oceanbase-ce-4.3.5.3
[2025-08-25 15:05:43.328] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - Call plugin oceanbase-ce-py_script_start-3.1.0 for oceanbase-ce-4.3.5.3-103000092025080818.el7-51410153665c32c5e0c66a4cb1cbe94648da12fb
[2025-08-25 15:05:43.328] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - import start
[2025-08-25 15:05:43.329] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - add start ref count to 1
[2025-08-25 15:05:43.330] [e55967dc-8181-11f0-81e6-707990b8c776] [INFO] Start observer
[2025-08-25 15:05:43.331] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- starting 127.0.0.1 observer
[2025-08-25 15:05:43.334] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- root@127.0.0.1 export LD_LIBRARY_PATH='/root/oceanbase-jamy/lib:'
[2025-08-25 15:05:43.334] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- local execute: cd /root/oceanbase-jamy; /root/oceanbase-jamy/bin/observer -r '127.0.0.1:2889:2888' -p 2888 -P 2889 -z 'zone1' -n 'oceanbase-jamy' -c 1756104405 -d '/root/oceanbase-jamy/store' -i 'lo' -o __min_full_resource_pool_memory=1073741824,memory_limit='10G',system_memory='3G',enable_syslog_recycle=True,enable_syslog_wf=False,max_syslog_file_count=4,cpu_count=2,datafile_size='5G',log_disk_size='30G',large_query_threshold='600s',enable_record_trace_log=False
[2025-08-25 15:05:44.009] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- exited code -4, error output:
[2025-08-25 15:05:44.009] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG]
[2025-08-25 15:05:44.009] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] -- root@127.0.0.1 export LD_LIBRARY_PATH=''
[2025-08-25 15:05:44.116] [e55967dc-8181-11f0-81e6-707990b8c776] [ERROR] [ERROR] OBD-2002: Failed to start 127.0.0.1 observer:
[2025-08-25 15:05:44.117] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - sub start ref count to 0
[2025-08-25 15:05:44.117] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - export start
[2025-08-25 15:05:44.117] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - plugin oceanbase-ce-py_script_start-3.1.0 result: None
[2025-08-25 15:05:44.124] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - exclusive lock /root/.obd/lock/mirror_and_repo release, count 2
[2025-08-25 15:05:44.125] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - exclusive lock /root/.obd/lock/mirror_and_repo release, count 1
[2025-08-25 15:05:44.125] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - exclusive lock /root/.obd/lock/mirror_and_repo release, count 0
[2025-08-25 15:05:44.125] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - try to get share lock /root/.obd/lock/mirror_and_repo
[2025-08-25 15:05:44.125] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - share lock /root/.obd/lock/mirror_and_repo release, count 2
[2025-08-25 15:05:44.125] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - exclusive lock /root/.obd/lock/deploy_oceanbase-jamy release, count 1
[2025-08-25 15:05:44.125] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - share lock /root/.obd/lock/mirror_and_repo release, count 1
[2025-08-25 15:05:44.126] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - share lock /root/.obd/lock/mirror_and_repo release, count 0
[2025-08-25 15:05:44.126] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - unlock /root/.obd/lock/mirror_and_repo
[2025-08-25 15:05:44.126] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - exclusive lock /root/.obd/lock/deploy_oceanbase-jamy release, count 0
[2025-08-25 15:05:44.126] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - unlock /root/.obd/lock/deploy_oceanbase-jamy
[2025-08-25 15:05:44.126] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - share lock /root/.obd/lock/global release, count 0
[2025-08-25 15:05:44.126] [e55967dc-8181-11f0-81e6-707990b8c776] [DEBUG] - unlock /root/.obd/lock/global
[2025-08-25 15:05:44.127] [e55967dc-8181-11f0-81e6-707990b8c776] [INFO] See https://www.oceanbase.com/product/ob-deployer/error-codes .
[2025-08-25 15:05:44.127] [e55967dc-8181-11f0-81e6-707990b8c776] [INFO] Trace ID: e55967dc-8181-11f0-81e6-707990b8c776
[2025-08-25 15:05:44.127] [e55967dc-8181-11f0-81e6-707990b8c776] [INFO] If you want to view detailed obd logs, please run: obd disp
#21

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/root/oceanbase-jamy/lib/
你这个太奇怪了。正常情况启动ob就会有日志输出，试试 /root/oceanbase-jamy/bin/observer

#22

你好，修改了 LD_LIBRARY_PATH 重新部署之后也没有输出日志。
observer貌似启动失败了。

[root@localhost log]# /root/oceanbase-jamy/bin/observer
Illegal instruction (core dumped)