oms 4.2.10_ce 增量同步 源 oceanbase 4.3.5.3 报错

【 使用环境 】生产环境
【 OB or 其他组件 】oms
【 使用版本 】oms 4.2.10_ce 源 oceanbase 4.3.5.3 目标 oceanbase 4.5.0.0
【问题描述】增量同步启动失败，报错
【复现路径】简单同步几张表，增量启动失败
【附件及日志】

错误日志
[2026-01-21 13:55:50.061968] INFO wait (ob_timer_service.cpp:271) [122963][][T0][Y0-0000000000000000-0-0] [lt=9] ObTimerService wait success(tenant_id=500, this=0x7f0e06370600)
992
[2026-01-21 13:55:50.061994] INFO destroy (ob_timer_service.cpp:278) [122963][][T0][Y0-0000000000000000-0-0] [lt=20] ObTimerService destroyed(tenant_id=500, this=0x7f0e06370600)
993
[2026-01-21 13:55:50.062002] INFO destroy_tg (thread_mgr.cpp:89) [122963][][T0][Y0-0000000000000000-0-0] [lt=6] destroy tg(tg_id=10, tg=0x7f0e1c7b3820, tg->attr_={name:qth_mgr, type:2})
994
[2026-01-21 13:55:50.260613] WDIAG [STORAGE.TRANS] getClock (ob_clock_generator.h:70) [122966][qth_mgr][T0][Y0-0000000000000000-0-0] [lt=39][errcode=-4006] clock generator not inited
995
[2026-01-21 13:55:50.260782] INFO [TLOG] do_destroy_ (ob_log_instance.cpp:1490) [122963][][T0][Y0-0000000000000000-0-0] [lt=10] destroy obcdc end
996
[2026-01-21 13:55:50.260807] EDIAG [TLOG] init_common_ (ob_log_instance.cpp:609) [122963][][T0][Y0-0000000000000000-0-0] [lt=18][errcode=-4002] init_components_ fail(ret=-4002, ret=“OB_INVALID_ARGUMENT”, start_tstamp_ns=1768974048000000000) BACKTRACE:0x154cb06b 0x152213c5 0x15396dcf 0x15396796 0x153966c8 0x153964e6 0x1a0f7bf0 0x1a0ec418 0x1a0ecd2e 0x1a0ecaa9 0x7f0e301abda3 0x7f0e3c457952 0x7f0e3bddbea5 0x7f0e3b3f5b0d
997
[2026-01-21 13:55:50.260882] EDIAG [TLOG] init (ob_log_instance.cpp:260) [122963][][T0][Y0-0000000000000000-0-0] [lt=74][errcode=-4002] init fail(ret=-4002, ret=“OB_INVALID_ARGUMENT”, start_tstamp_usec=1768974048000000) BACKTRACE:0x154cb06b 0x152213c5 0x15396dcf 0x15396796 0x153966c8 0x153964e6 0x1a0ecbdc 0x1a0ecad3 0x7f0e301abda3 0x7f0e3c457952 0x7f0e3bddbea5 0x7f0e3b3f5b0d
998
[2026-01-21 13:55:50.260990] INFO [TLOG] destroy_instance (ob_log_instance.cpp:126) [122963][][T0][Y0-0000000000000000-0-0] [lt=36] ObLogInstance 0x7f0e300f9010 destroy
999
[2026-01-21 13:55:50.260997] INFO [TLOG] ~ObLogInstance (ob_log_instance.cpp:213) [122963][][T0][Y0-0000000000000000-0-0] [lt=6] ====================libobcdc end====================
1000
[2026-01-21 13:55:50.261465] INFO destroy_array (ob_dchash.h:427) [122963][][T0][Y0-0000000000000000-0-0] [lt=5] DCHash: destroy_array: N9oceanbase6common9sqlclient12TenantMapKeyE this=0x7f0e3011bfd0 array=0x7f0d71a04030 array_size=65536

【备注】基于 LLM 和开源文档 RAG 的论坛小助手已开放测试，在发帖时输入 [@论坛小助手] 即可召唤小助手，欢迎试用！

oms监控组件截图看看 如果是增加组件有问题 麻烦发一下日志信息

异常日志和输出日志增量的libobcdc.log

稍等，刚才配置的任务被我删了，我来复现一下

[2026-01-21 15:29:49.196838] WDIAG [STORAGE.TRANS] getClock (ob_clock_generator.h:70) [150335][TimerWK1][T0][Y0-0000000000000000-0-0] [lt=32][errcode=-4006] clock generator not inited
646
[2026-01-21 15:29:49.196843] WDIAG [STORAGE.TRANS] getClock (ob_clock_generator.h:70) [150334][TimerWK0][T0][Y0-0000000000000000-0-0] [lt=27][errcode=-4006] clock generator not inited
647
[2026-01-21 15:29:49.196848] WDIAG [STORAGE.TRANS] getClock (ob_clock_generator.h:70) [150337][TimerWK3][T0][Y0-0000000000000000-0-0] [lt=23][errcode=-4006] clock generator not inited
648
[2026-01-21 15:29:49.196860] WDIAG [STORAGE.TRANS] getClock (ob_clock_generator.h:70) [150336][TimerWK2][T0][Y0-0000000000000000-0-0] [lt=15][errcode=-4006] clock generator not inited
649
[2026-01-21 15:29:49.197119] INFO wait (ob_timer_service.cpp:271) [150330][][T0][Y0-0000000000000000-0-0] [lt=9] ObTimerService wait success(tenant_id=500, this=0x7fd7ea370600)
650
[2026-01-21 15:29:49.197146] INFO destroy (ob_timer_service.cpp:278) [150330][][T0][Y0-0000000000000000-0-0] [lt=20] ObTimerService destroyed(tenant_id=500, this=0x7fd7ea370600)
651
[2026-01-21 15:29:49.197155] INFO destroy_tg (thread_mgr.cpp:89) [150330][][T0][Y0-0000000000000000-0-0] [lt=7] destroy tg(tg_id=10, tg=0x7fd8001b3820, tg->attr_={name:qth_mgr, type:2})
652
[2026-01-21 15:29:49.197329] WDIAG [STORAGE.TRANS] getClock (ob_clock_generator.h:70) [150333][qth_mgr][T0][Y0-0000000000000000-0-0] [lt=25][errcode=-4006] clock generator not inited
653
[2026-01-21 15:29:49.197477] INFO [TLOG] do_destroy_ (ob_log_instance.cpp:1490) [150330][][T0][Y0-0000000000000000-0-0] [lt=12] destroy obcdc end
654
[2026-01-21 15:29:49.197495] EDIAG [TLOG] init_common_ (ob_log_instance.cpp:609) [150330][][T0][Y0-0000000000000000-0-0] [lt=11][errcode=-4002] init_components_ fail(ret=-4002, ret=“OB_INVALID_ARGUMENT”, start_tstamp_ns=1768979836000000000) BACKTRACE:0x154cb06b 0x152213c5 0x15396dcf 0x15396796 0x153966c8 0x153964e6 0x1a0f7bf0 0x1a0ec418 0x1a0ecd2e 0x1a0ecaa9 0x7fd81426dda3 0x7fd81e516952 0x7fd81de9aea5 0x7fd81d4b4b0d
655
[2026-01-21 15:29:49.197578] EDIAG [TLOG] init (ob_log_instance.cpp:260) [150330][][T0][Y0-0000000000000000-0-0] [lt=82][errcode=-4002] init fail(ret=-4002, ret=“OB_INVALID_ARGUMENT”, start_tstamp_usec=1768979836000000) BACKTRACE:0x154cb06b 0x152213c5 0x15396dcf 0x15396796 0x153966c8 0x153964e6 0x1a0ecbdc 0x1a0ecad3 0x7fd81426dda3 0x7fd81e516952 0x7fd81de9aea5 0x7fd81d4b4b0d
656
[2026-01-21 15:29:49.197764] INFO [TLOG] destroy_instance (ob_log_instance.cpp:126) [150330][][T0][Y0-0000000000000000-0-0] [lt=24] ObLogInstance 0x7fd80c51f010 destroy
657
[2026-01-21 15:29:49.197777] INFO [TLOG] ~ObLogInstance (ob_log_instance.cpp:213) [150330][][T0][Y0-0000000000000000-0-0] [lt=11] ====================libobcdc end====================
658
[2026-01-21 15:29:49.198329] INFO destroy_array (ob_dchash.h:427) [150330][][T0][Y0-0000000000000000-0-0] [lt=5] DCHash: destroy_array: N9oceanbase6common9sqlclient12TenantMapKeyE this=0x7fd80c541fd0 array=0x7fd750004030 array_size=65536

按照我楼上发的 日志信息 发一下

这就是你说的日志 我没法上传log文件，复制了一段 libobcdc.log 里面内容 在上面的回复中

尽量全部都提供一下 还有这个路径下的日志也提供一下

看着有这个报错 建议检查ob数据库 启动是否正常

oms容器内执行 ll /home/ds/lib64/reader/ob-ce-4.3-reader/ 发一下

业务一直在用的数据库，只是想配置这个库增量同步几个表到另一个库。咋可能ob都没启动

系统租户oceanbase库查询下这个，发一下结果

WITH palf_log_stat AS (
SELECT
tenant_id,
MAX(begin_scn) AS palf_available_start_scn,
MIN(end_scn) AS palf_available_latest_scn,
SCN_TO_TIMESTAMP(MAX(begin_scn)) AS palf_available_start_scn_display,
SCN_TO_TIMESTAMP(MIN(end_scn)) AS palf_available_latest_scn_display
FROM GV$OB_LOG_STAT
WHERE tenant_id & 0x01 = 0 or tenant_id = 1
GROUP BY tenant_id
),
archivelog_stat AS (
SELECT
a.tenant_id AS tenant_id,
MIN(b.start_scn) AS archive_start_scn,
a.checkpoint_scn AS archive_latest_scn,
a.checkpoint_scn_display AS archive_available_latest_scn_display
FROM CDB_OB_ARCHIVELOG a
LEFT JOIN CDB_OB_ARCHIVELOG_PIECE_FILES b
ON a.tenant_id = b.tenant_id AND a.round_id = b.round_id
AND b.file_status != ‘DELETED’ AND a.STATUS = ‘DOING’
GROUP BY a.tenant_id
)
SELECT
pls.tenant_id,
pls.palf_available_start_scn,
pls.palf_available_latest_scn,
pls.palf_available_start_scn_display AS palf_available_start_scn_display,
pls.palf_available_latest_scn_display AS palf_available_latest_scn_display,
als.archive_start_scn AS archive_available_start_scn,
als.archive_latest_scn AS archive_available_latest_scn,
CASE WHEN als.archive_start_scn IS NOT NULL THEN SCN_TO_TIMESTAMP(als.archive_start_scn) ELSE NULL END AS archive_available_start_scn_dispalay,
als.archive_available_latest_scn_display
FROM palf_log_stat pls
LEFT JOIN archivelog_stat als ON pls.tenant_id = als.tenant_id
GROUP BY pls.tenant_id, pls.palf_available_start_scn;

1002 是我用的租户

mysql的很正常，ob的库一直卡在这里
日志里能看到感觉有用的错误日志

[2026-01-23 14:46:17.518701] INFO [TLOG] query (ob_cdc_tenant_query.h:117) [155444][][T0][Y0-0000000000000000-0-0] [lt=46] tenant query retring(ret=-2002, ret=“Unknown Error”, tenant_id=1002, retry_cnt=451, remain_retry_time=14959317, sql=SELECT SNAPSHOT_SCN, ORA_ROWSCN AS END_SCN, START_LSN, END_LSN FROM __all_data_dictionary_in_log WHERE SNAPSHOT_SCN <= 1769149641000000000 ORDER BY SNAPSHOT_SCN DESC LIMIT 1)

我把这个sql在我的租户里去执行 SELECT SNAPSHOT_SCN, ORA_ROWSCN AS END_SCN, START_LSN, END_LSN FROM oceanbase.__all_data_dictionary_in_log ORDER BY SNAPSHOT_SCN DESC LIMIT 1; 返回：

1769150875054461000 1769150875116113000 193893584 193893584

这个任务的创建时间是多少

社区里我搜到一个类似的帖子，oms数据迁移失败 - #20，来自 三木大大

我怀疑是ip的问题

oceanbase-ce:
servers:
# Please don’t use hostname, only IP can be supported

• 127.0.0.1

单机部署写成本地地址了，等晚上改为内网地址重启数据库后再试。

看到报错 只是怀疑 但是如果使用了 * 127.0.0.1 是不推荐的 可以改一下地址 在试一下