Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: RDS实例主备故障切换,DM找不对应的GTID导致中断
[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version] 6.1.1
[Encountered Problem] RDS instance switch, DM cannot find the corresponding GTID causing interruption
[Reproduction Path]
[Problem Phenomenon and Impact]
RDS instance master-slave failover, it should be an issue with RDS, DM cannot find the corresponding GTID causing interruption, has anyone encountered this?
[Attachment]
Please provide the version information of each component, such as cdc/tikv, which can be obtained by executing cdc version/tikv-server --version.
The image you provided is not accessible. Please provide the text content that needs to be translated.
Change DM to GTID mode for synchronization, and make the adjustments again. T-Mac
This needs to be resynchronized.
Set GTID from the beginning.
Are you asking if the enable_gtid
in the DM source configuration file should be set to true?
You can first supplement the configuration file, and then follow the prompts.
Please check if the binlog file could be parsed by mysqlbinlog
.
Use mysqlbinlog to check the corresponding binlog file and see its contents. Then continue with the subsequent troubleshooting.
It should be that DM synchronization is fast. After the RDS master-slave switch, the GTID of the new master has not yet reached the GTID synchronized by DM. You can try specifying the GTID again.
The synchronized instance is a slave read-only instance, with no binlog download, and skipping the relevant GTID also reports an error.
How to synchronize without binlog?
RDS does not provide binlog downloads for slave read-only. This issue was reported to Alibaba support as a bug.
Did he not enable binlog? DM cannot be used if the upstream does not enable binlog. You can use check-task to check.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.